Sensory feedback can give rise to neural rotations

Investigating how an artificial network of neurons controls a simulated arm suggests that rotational patterns of activity in the motor cortex may rely on sensory feedback from the moving limb.

Position Estimation at Zero Speed for PMSMs Using Artificial Neural Networks

This paper presents a method for shaft position estimation of a synchronous motor with permanent magnets. Zero speed and very low speed range are considered. The method uses the analysis of high-frequency currents induced by the introduction of additional voltage in the control path in the stationary coordinate system associated with the stator. An artificial neural network estimates the sine and cosine values necessary in the Park’s transformation units. This method can achieve satisfactory accuracy in the case of low asymmetry of inductance in the direct and quadrature axes of the coordinate system associated with the rotor. The TensorFlow/Keras package was used for artificial network calculations and the scikit-learn package for preprocessing. Aggregating the outputs of several artificial neural networks provides an opportunity to reduce the resultant estimation error. The use of as few as four networks has enabled the error to be reduced by approximately 20% compared to a single example network.

Speech Emotion Recognition

Speech Emotion Recognition, abbreviated as SER, the act of trying to identify a person's feelings and relationships. Affected situations from speech. This is because the truth often reflects the basic feelings of tone and tone of voice. Emotional awareness is a fast-growing field of research in recent years. Unlike humans, machines do not have the power to comprehend and express emotions. But human communication with the computer can be improved by using automatic sensory recognition, accordingly reducing the need for human intervention. In this project, basic emotions such as peace, happiness, fear, disgust, etc. are analyzed signs of emotional expression. We use machine learning techniques such as Multilayer perceptron Classifier (MLP Classifier) which is used to separate information provided by groups to be divided equally. Coefficients of Mel-frequency cepstrum (MFCC), chroma and mel features are extracted from speech signals and used to train MLP differentiation. By accomplishing this purpose, we use python libraries such as Librosa, sklearn, pyaudio, numpy and audio file to analyze speech patterns and see the feeling. Keywords: Speech emotion recognition, mel cepstral coefficient, neural artificial network, multilayer perceptrons, mlp classifier, python.

Artificial Intelligence- (AI-) Enabled Internet of Things (IoT) for Secure Big Data Processing in Multihoming Networks

The automated techniques enabled with Artificial Neural Networks (ANN), Internet of Things (IoT), and cloud-based services affect the real-time analysis and processing of information in a variety of applications. In addition, multihoming is a type of network that combines various types of networks into a single environment while managing a huge amount of data. Nowadays, the big data processing and monitoring in multihoming networks provide less attention while reducing the security risk and efficiency during processing or monitoring the information. The use of AI-based systems in multihoming big data with IoT- and AI-integrated systems may benefit in various aspects. Although multihoming security issues and their analysis have been well studied by various scientists and researchers; however, not much attention is paid towards big data security processing in multihoming especially using automated techniques and systems. The aim of this paper is to propose an IoT-based artificial network to process and compute big data processing by ensuring a secure communication multihoming network using the Bayesian Rule (BR) and Levenberg-Marquardt (LM) algorithms. Further, the efficiency and effect on multihoming information processing using an AI-assisted mechanism are experimented over various parameters such as classification accuracy, classification time, specificity, sensitivity, ROC, and F -measure.

Framework based on parameterized images on ResNet to identify intrusions in smartwatches or other related devices

A conceptual and algebraically framework that did not exist up until then in its morphology, was developed. What is more, it is pioneer on its implementation in the area of Artificial Intelligence (AI) and it was started up in laboratory, on its structural aspects, as a fully operational model. At qualitative level, its greatest contribution to AI is applying the conversion or transduction of parameters obtained by ternary logic (multi-valued systems) and associating them with an image. This image will be analysed by means of a residual artificial network ResNet34, to warn us of an intrusion. The field of application of this framework includes everything from smartwatches, tablets, and PC’s to the home automation based on the KNX standard.

Smartphone Sensor Accelerometer Data for Human Activity Recognition Using Spiking Neural Network

Recognizing human actions is a challenging task and actively research in computer vision community. The task of human activity recognition has been widely used in various application such as human monitoring in a hospital or public spaces. This work applied open dataset of smartphones accelerometer data for various type of activities. The analogue input data is encoded into the spike trains using some form of a rate-based method. Spiking neural network is a simplified form of dynamic artificial network. Therefore, this network is expected to model and generate action potential from the leaky integrate-and-fire spike response model. The leaning rule is adaptive and efficient to present synapse exciting and inhibiting firing neuron. The result found that the proposed model presents the state-of-the-art performance at a low computational cost.

PRONÓSTICO ANUAL DE LA CARGA TÉRMICA RADIANTE APLICÁNDOSE INTELIGENCIA ARTIFICIAL

En este estudio se desenvolvieron redes neuronales artificiales para predecir el conforto térmico animal, en función de la temperatura ambiente y la velocidad del aire para cada día del año en el calendario juliano. Los datos fueron obtenidos en el sitio del Instituto Nacional de Meteorología para una serie histórica de 30 años, coleccionada en la Estación Convencional Padre Ricardo Remetter, municipio de Santo Antonio de Leverger-MT. Para la elaboración de las redes se adoptó como variable de entrada el día del año y como variable de salida la carga térmica de radiación. El número de neuronas varió entre 2 y 15, utilizándose una y dos camadas ocultas. El ajuste de las redes se verificó por el coeficiente de determinación, error absoluto medio, porcentaje medio del error absoluto, la normalidad de los residuos y la prueba de t-Student. No hubo discrepancias entre los valores estimados por las redes y los obtenidos de la serie histórica. Finalmente se seleccionaron diez arquitecturas con adecuados índices de desempeño y las cuatro mejores se sometieron al análisis de residuos. Se concluyó que las redes neuronales del tipo perceptron con dos camadas ocultas fueron apropiadas para pronosticar la carga térmica radiante, conforme el día Juliano. Palabras-clave: conforto térmico; red neuronal artificial; series temporales. Annual prognostic of the radiant thermal using artificial intelligence ABSTRACT: In this research, artificial neural networks were developed to predict the animal thermal comfort based on the room temperature and air velocity for the year day in the Julian calendar. The data were obtained from the website of the National Institute of Meteorology for a 30-year historical series, collected at the Padre Ricardo Remetter Meteorological Station, municipality of Santo Antônio de Leverger-MT. To elaborate the networks, the day of the year was adopted as the input variable and the radiation thermal load as the output variable. The number of neurons ranged varied from 2 to 15, being used one and two hidden layers. The adjustment of the networks was verified by the determination coefficient, mean absolute error, mean percentage of the absolute error, the normality of residues and the t-Student test. The values estimated by the networks and those obtained from the historical series did not differ. Finally, ten architectures with adequate performance and efficiency indexes were selected and among them the four best were submitted to the residue analysis. It was concluded that the artificial perceptron neural networks formed by two-layer hidden were suitable for the prognosis of the radiant thermal load, as a function of Julian day. Keywords: thermal comfort; artificial network; time series.