Trajectory Tracking Control for Underactuated USV with Prescribed Performance and Input Quantization

This paper is devoted to the problem of prescribed performance trajectory tracking control for symmetrical underactuated unmanned surface vessels (USVs) in the presence of model uncertainties and input quantization. By combining backstepping filter mechanisms and adaptive algorithms, two robust control architectures are investigated for surge motion and yaw motion. To guarantee the prespecified performance requirements for position tracking control, the constrained error dynamics are transformed to unconstrained ones by virtue of a tangent-type nonlinear mapping function. On the other hand, the inaccurate model can be identified through radial basis neural networks (RBFNNs), where the minimum learning parameter (MLP) algorithm is employed with a low computational complexity. Furthermore, quantization errors can be effectively reduced even when the parameters of the quantizer remain unavailable to designers. Finally, the effectiveness of the proposed controllers is verified via theoretical analyses and numerical simulations.

Anti-swing radial basis neuro-fuzzy linear quadratic regulator control of double link rotary pendulum

In this article, a radial basis neuro-fuzzy linear quadratic regulator controller is developed for the anti-swing control of a double link rotary pendulum system. The objective of this work is to study the radial basis neuro-fuzzy linear quadratic regulator controller and to compare it with a fuzzy linear quadratic regulator and the linear quadratic regulator controllers. In the proposed radial basis neuro-fuzzy linear quadratic regulator controllers, the positions and velocities of state variables multiplied by their linear quadratic regulator gains are trained using two radial basis neural networks architecture. The output of the two radial basis neural networks is used as the input variables of the fuzzy controller. The novel architecture of the radial basis neuro-fuzzy controller is developed in order to obtain better control performance than the classical adaptive neuro-fuzzy controller. To determine the control performance of the anti-swing controllers, different control parameters are computed. According to the comparative results, the anti-swing radial basis neuro-fuzzy linear quadratic regulator controller yields improved results than fuzzy linear quadratic regulator and linear quadratic regulator. Furthermore, the performance of the three controllers developed was compared based on robustness analysis under external force disturbance. The results obtained here indicate that the anti-swing radial basis neuro-fuzzy linear quadratic regulator controller product has better performance than other controllers in terms of vibration suppression ability.

Classification of the Microstructural Elements of the Vegetal Tissue of the Pumpkin (Cucurbita pepo L.) Using Convolutional Neural Networks

Although knowledge of the microstructure of food of vegetal origin helps us to understand the behavior of food materials, the variability in the microstructural elements complicates this analysis. In this regard, the construction of learning models that represent the actual microstructures of the tissue is important to extract relevant information and advance in the comprehension of such behavior. Consequently, the objective of this research is to compare two machine learning techniques—Convolutional Neural Networks (CNN) and Radial Basis Neural Networks (RBNN)—when used to enhance its microstructural analysis. Two main contributions can be highlighted from this research. First, a method is proposed to automatically analyze the microstructural elements of vegetal tissue; and second, a comparison was conducted to select a classifier to discriminate between tissue structures. For the comparison, a database of microstructural elements images was obtained from pumpkin (Cucurbita pepo L.) micrographs. Two classifiers were implemented using CNN and RBNN, and statistical performance metrics were computed using a 5-fold cross-validation scheme. This process was repeated one hundred times with a random selection of images in each repetition. The comparison showed that the classifiers based on CNN produced a better fit, obtaining F1–score average of 89.42% in front of 83.83% for RBNN. In this study, the performance of classifiers based on CNN was significantly higher compared to those based on RBNN in the discrimination of microstructural elements of vegetable foods.

Verification the Related Data Base by the Layered Radial-Basis Neural Network

The verification method of the information in the related data bases by the theory of layered radial-basis neural networks given in this article. The method consists of three levels: clustering the data by Kohonen network, learning and usage of radial-basis neuron network, marking the reliability of the rows in the table of data base. The software worked out by the invers method in the basis of radial-neuron basis. The suggested data base by the method of verification gives opportunity to work out the models of integrated informational systems, algorithms and software.