Current Issue : October - December Volume : 2020 Issue Number : 4 Articles : 5 Articles
This paper proposes an adaptive predefined performance neural control scheme for robotic manipulators in the presence of\nnonlinear dead zone. A neural network (NN) is utilized to estimate the model uncertainties and unknown dynamics. An improved\nfunnel function is designed to guarantee the transient behavior of the tracking error. The proposed funnel function can release the\nassumption on the conventional funnel control. Then, an adaptive predefined performance neural controller is proposed for\nrobotic manipulators, while the tracking errors fall within a prescribed funnel boundary.Theclosed-loop system stability is proved\nvia Lyapunov function. Finally, the numerical simulation results based on a 2-DOF robotic manipulator illustrate the control effect\nof the presented approach....
To solve the problems of high complexity and low accuracy in Volterra time-domain kernel calculation of a nonlinear system, this\npaper proposes an intelligent calculation method of Volterra time-domain kernel by time-delay artificial neural networks\n(TDANNs) and also designs a root mean square error (RMSE) index to choose the neuron number of the network input layer.\nFirstly, a three-layer TDANN is designed according to the characteristics of the Volterra model. Secondly, the relationship\nbetween parameters of TDANN and Volterra time-domain kernel is analyzed, and then three-order expressions of Volterra timedomain\nkernel are derived. The calculation of Volterra time-domain kernel is completed by network training. Finally, it is verified\nby a nonlinear system. Simulation results indicate that compared with traditional methods, the new method has higher accuracy,\nand it can realize the batch calculation of Volterra kernel, which not only improves the calculation efficiency but also provides\naccurate data for fault diagnosis based on Volterra kernel in further research work....
Reducing the dimension of the hyperspectral image data can directly reduce the redundancy of the data, thus improving the\naccuracy of hyperspectral image classification. In this paper, the deep belief network algorithm in the theory of deep learning is\nintroduced to extract the in-depth features of the imaging spectral image data. Firstly, the original data is mapped to feature space\nby unsupervised learning methods through the Restricted Boltzmann Machine (RBM). Then, a deep belief network will be formed\nby superimposed multiple Restricted Boltzmann Machines and training the model parameters by using the greedy algorithm layer\nby layer. At the same time, as the objective of data dimensionality reduction is achieved, the underground feature construction of\nthe original data will be formed. The final step is to connect the depth features of the output to the Softmax regression classifier to\ncomplete the fine-tuning (FT) of the model and the final classification. Experiments using imaging spectral data showing the indepth\nfeatures extracted by the profound belief network algorithm have better robustness and separability. It can significantly\nimprove the classification accuracy and has a good application prospect in hyperspectral image information extraction....
An efficient storage strategy for retail e-commerce warehousing is important for minimizing the order retrieval time to improve\nthe warehouse-output efficiency. In this paper, we consider a model and algorithm to solve the cargo location problem in a retail\ne-commerce warehouse. The problem is abstracted into storing cargo on three-dimensional shelves, and the mathematical model\nis built considering three objectives: efficiency, stability, and classification. An artificial swarm algorithm is designed to solve the\nproposed models. Computational experiments performed on a warehouse show that the proposed approach is effective at solving\nthe cargo location assignment problem and is significant for the operation and organization of a retail e-commerce warehouse....
The importance of eating behavior risk factors in the primary prevention of obesity has been established. Researchers mostly use\nthe linear model to determine associations among these risk factors. However, in reality, the presence of nonlinearity among these\nfactors causes a bias in the prediction models. The aim of this study was to explore the potential of a hybrid model to predict the\neating behaviors. The hybrid model of structural equation modelling (SEM) and artificial neural networks (ANN) was applied to\nevaluate the prediction model. The SEM analysis was used to check the relationship of the emotional eating scale (EES), body\nshape concern (BSC), and body appreciation scale (BAS) and their effect on different categories of eating behavior patterns (EBP).\nIn the second step, the input and output required for ANN analysis were obtained from SEM analysis and were applied in the\nneural network model. 340 university students participated in this study. The hybrid model (SEM-ANN) was conducted using\nmultilayer perceptron (MLP) with feed-forward network topology. Moreover, Levenbergâ??Marquardt, which is a supervised\nlearning model, was applied as a learning method for MLP training. The tangent/sigmoid function was used for the input layer,\nwhile the linear function was applied for the output layer. The coefficient of determination (R2) and mean square error (MSE) were\ncalculated. Using the hybrid model, the optimal network happened at MLP 3-17-8. It was proved that the hybrid model was\nsuperior to SEM methods because the R2 of the model was increased by 27%, while the MSE was decreased by 9.6%. Moreover, it\nwas found that BSC, BAS, and EES significantly affected healthy and unhealthy eating behavior patterns. Thus, a hybrid approach\ncould be suggested as a significant methodological contribution from a machine learning standpoint, and it can be implemented as\nsoftware to predict models with the highest accuracy....
Loading....