scholarly journals Kendali Model Kursi Roda dengan Electromyograf dan Accelerometer Menggunakan Metode Jaringan Saraf Tiruan

2020 ◽  
Vol 1 (2) ◽  
pp. 59-68
Author(s):  
Adib Nur Sasongko
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Control System ◽  
Muscle Activity ◽  
People With Disabilities ◽  
Muscle Contractions ◽  
Human Muscle ◽  
Accelerometer Sensor ◽  
Actuator System ◽  
Arduino Uno

Kursi roda digunakan sebagai alat bantu untuk berpindah tempat bagi orang yang memiliki disabilitas pada bagian kaki. Namun, kursi roda konvensional biasanya dijalankan dengan mendorong kursi roda tersebut atau pengguna mengendalikan sendiri roda dari kursi roda yang dikendarainya. Namun untuk pengguna yang juga memiliki disabilitas pada tangannya tidak dapat mengendalikan sendiri kursi roda konvensional. Sebuah robot kursi roda dirancang untuk memudahkan para pengguna yang memiliki disabilitas khusus pada kaki dan tangannya. Robot ini dirancang untuk dapat dijalankan dengan menggerakkan otot pada dahi dan dengan cara memiringkan kepala. Jaringan saraf tiruan digunakan untuk memprediksi nilai sensor selanjutnya dan membandingkan dengan data nilai sensor sebelumnya untuk menentukan adanya kontraksi pada otot. Robot kursi roda ini dibuat dengan tiga bagian utama yaitu masukan dengan menggunakan sensor EMG (Electromyograf) dan sensor Accelerometer, sistem pengendali berbasis Arduino UNO dan sistem aktuator dengan menggunakan motor DC. Electromyograf (EMG) merupakan peralatan yang digunakan untuk menangkap aktivitas otot manusia yang pada penelitian ini sensor EMG dipasang pada dahi. Accelerometer yang nantinya akan dipasang pada bagian kepala mempunyai fungsi untuk mengukur percepatan akibat gravitasi bumi sehingga dapat digunakan untuk menentukan arah yang akan dituju. Pada penelitian ini hasil yang didapatkan adalah model mampu mendeteksi kontraksi otot dan dapat menentukan arah kursi roda sesuai yang diinginkan. Wheelchairs are used as a tool to move places for people with disabilities in their legs. However, conventional wheelchairs are usually run by pushing the wheelchair or the user controls the wheel himself from the wheelchair he is driving. However, for users who also have disabilities in their hands, they cannot control conventional wheelchairs themselves. A wheelchair robot is designed to make it easier for users who have special disabilities on their feet and hands. This robot is designed to be able to run by moving the muscles on the forehead and by tilting the head. Artificial neural networks are used to predict subsequent sensor values ​​and compare with previous sensor value data to determine the presence of muscle contractions. This wheelchair robot is made with three main parts namely input using an EMG (Electromyograph) sensor and an Accelerometer sensor, a control system based on Arduino UNO and an actuator system using a DC motor. Electromyograph (EMG) is an equipment used to capture human muscle activity which in this study EMG sensors were installed on the forehead. Accelerometer which will be installed on the head has a function to measure the acceleration due to gravity so that it can be used to determine the direction to be headed. In this study the results obtained are the model can detect muscle contractions and can determine the direction of the wheelchair as desired.

scholarly journals Methodology for Ensuring a Comfortable Microclimate State in a Smart Home Using an Ensemble of Fuzzy Artificial Neural Networks

2021 ◽  
Vol 20 (6) ◽  
pp. 1418-1447
Author(s):  
Andrey Mozohin
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Control System ◽  
Physical Environment ◽  
Air Conditioning ◽  
Smart Home ◽  
Practical Implementation ◽  
Climate Control ◽  
Artificial Neural ◽  
Hybrid Neural Networks

Analysis of the application of smart home technology indicates an insufficient level of controllability of its infrastructure, which leads to excessive consumption of energy and information resources. The problem of managing the digital infrastructure of human living space, is associated with a large number of highly specialized solutions for home automation, which complicate the management process. Smart home is considered as a set of independent cyber-physical devices aimed at achieving its goal. For coordinated work of cyber-physical devices it is proposed to provide their joint work through a single information center. Simulation of device operation modes in a digital environment preserves the resource of physical devices by making a virtual calculation for all possible variants of interaction of devices between themselves and the physical environment. A methodology for controlling the microclimate of a smart home using an ensemble of fuzzy artificial neural networks is developed, with the example of joint use of air conditioning, ventilation and heating. The neural network algorithm allows you to monitor the parameters of the physical environment, predict the modes of cyber-physical devices and generate control signals for each of them, ensuring the joint operation of devices with minimal resource consumption and information traffic. A variant of practical implementation of a smart home climate control system on the example of a multifunctional educational computer class is proposed. Hybrid neural networks of air conditioning, ventilation and heating systems were developed. The testing of the microclimate control system of a multifunctional university classroom using hybrid neural networks was carried out, a programmable logic controller of domestic production was used as a control device. The goal of management based on cooperating cyber-physical devices is to achieve a minimum of power and information traffic when they work together.

Neural Controlling of Remotely Operated Underwater Vehicle

2011 ◽  
Vol 180 ◽  
pp. 168-174 ◽  
Author(s):  
Andrzej Żak
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Control System ◽  
Underwater Vehicle ◽  
Mathematical Basis ◽  
Dynamical Object ◽  
Artificial Neural

The main aim of paper is to introduce the results of research concentrated on controlling remotely operated underwater vehicle using artificial neural networks. Firstly the mathematical basis of neural network using to control dynamical object were introduced. Next the proposed control system which is using technology of artificial neural network was presented. At the end the example results of research on stabilizing movements’ parameters of underwater vehicle using ROV simulator were presented. The paper is finished by summary which include conclusions derive from results of research.

Condition Monitoring of an Electrohydraulic Position Control System Using Artificial Neural Networks

2004 ◽  
Author(s):  
K. Pollmeier ◽  
C. R. Burrows ◽  
K. A. Edge
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Condition Monitoring ◽  
Position Control ◽  
Linear Actuator ◽  
Modular Approach ◽  
Failure Characteristics ◽  
Actuator System ◽  
Artificial Neural ◽  
Position Control System

This paper investigates the condition monitoring of a servo-valve-controlled linear actuator system using artificial neural networks (NNs). The aim is to discuss techniques for the identification of failure characteristics and their source. It is shown that neural networks can be trained to identify more than one fault but these are larger and require more training patterns than networks for single fault diagnosis. This leads to much longer training times and to problems with scaleability. Therefore a modular approach has been developed. Several networks were trained each to identify an individual fault. The parallel outputs of these nets were then used as inputs to another network. This additional network was able to identify not only the correct faults but also the actual fault levels.

pH-Control System Based on Artificial Neural Networks

1998 ◽  
Vol 37 (7) ◽  
pp. 2729-2740 ◽  
Author(s):  
María C. Palancar ◽  
José M. Aragón ◽  
José S. Torrecilla
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Control System ◽  
Ph Control ◽  
Artificial Neural
Sign in / Sign up

Export Citation Format

Share Document