Claim Missing Document
Check
Articles

Found 7 Documents
Search

Hybrid metaheuristic approach for robot path planning in dynamic environment Lina Basem Amar; Wesam M. Jasim
Bulletin of Electrical Engineering and Informatics Vol 10, No 4: August 2021
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/eei.v10i4.2836

Abstract

Recently robots have gained great attention due to their ability to operate in dynamic and complex environments with moving obstacles. The path planning of a moving robot in a dynamic environment is to find the shortest and safe possible path from the starting point towards the desired target point. A dynamic environment is a robot's environment that consists of some static and moving obstacles. Therefore, this problem can be considered as an optimization problem and thus it is solved via optimization algorithms. In this paper, three approaches for determining the optimal pathway of a robot in a dynamic environment were proposed. These approaches are; the particle swarming optimization (PSO), ant colony optimization (ACO), and hybrid PSO and ACO. These used to carry out the path planning tasks effectively. A set of certain constraints must be met simultaneously to achieve the goals; the shortest path, the least time, and free from collisions. The results are calculated for the two algorithms separately and then that of the hybrid algorithm is calculated. The effectiveness and superiority of the hybrid algorithm were verified on both PSO and ACO algorithms.
Pre-convoluted neural networks for fashion classification Mustafa Amer Obaid; Wesam M. Jasim
Bulletin of Electrical Engineering and Informatics Vol 10, No 2: April 2021
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/eei.v10i2.2750

Abstract

In this work, concept of the fashion-MNIST images classification constructed on convolutional neural networks is discussed. Whereas, 28×28 grayscale images of 70,000 fashion products from 10 classes, with 7,000 images per category, are in the fashion-MNIST dataset. There are 60,000 images in the training set and 10,000 images in the evaluation set. The data has been initially pre-processed for resizing and reducing the noise. Then, this data is normalized for ensuring that all the data are on the same scale and this usually improves the performance. After normalizing the data, it is augmented where one image will be in three forms of output. The first output image is obtained by rotating the actual one; the second output image is obtained as acute angle image; and the third is obtained as tilt image. The new data set is of 180,000 images for training phase and 30,000 images for the testing phase. Finally, data is sent to training process as input for training model of the pre-convolution network. The pre-convolution neural network with the five layered convoluted deep neural network and do the training with the augmented data, The performance of the proposed system shows 94% accuracy where it was 93% in VGG16 and 92% in AlexNetnetworks.
Diabetics blood glucose control based on GA-FOPID technique Wesam M. Jasim; Yousif I. Al Mashhadany
Bulletin of Electrical Engineering and Informatics Vol 10, No 1: February 2021
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/eei.v10i1.2715

Abstract

In this paper, an optimized Fractional Order Proportional, Integral, Derivative based Genetic Algorithm GA-FOPID optimization technique is proposed for glucose level normalization of diabetic patients. The insulin pump with diabetic patient system used in the simulation is the Bergman minimal model, which is used to simulate the overall system. The main purpose is to obtain the optimal controller parameters that regulate the system smoothly to the desired level using GA optimization to find the FOPID parameters. The next step is to obtain the FOPID controller parameters and the traditional PID controller parameters normally. Then, the simulation output results of using the proposed GA-FOPID controller was compared with that of using the normal FOPID and the traditional PID controllers. The comparison shows that all the three controllers can regulate the glucose level but the use of GA-FOPID controller was outperform the use of the other two controllers in terms of speed of normalization and the overshoot value.
Multi-objective optimization path planning with moving target Baraa M. Abed; Wesam M. Jasim
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 11, No 3: September 2022
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v11.i3.pp1184-1196

Abstract

Path planning or finding a collision-free path for mobile robots between starting position and its destination is a critical problem in robotics. This study is concerned with the multi objective optimization path planning problem of autonomous mobile robots with moving targets in dynamic environment, with three objectives considered: path security, length and smoothness. Three modules are presented in the study. The first module is to combine particle swarm optimization algorithm (PSO) with bat algorithm (BA). The purpose of PSO is to optimize two important parameters of BA algorithm to minimize distance and smooth the path. The second module is to convert the generated infeasible points into feasible ones using a new local search algorithm (LS). The third module obstacle detection and avoidance (ODA) algorithm is proposed to complete the path, which is triggered when the mobile robot detects obstacles in its field of vision. ODA algorithm based on simulating human walking in a dark room. Several simulations with varying scenarios are run to test the validity of the proposed solution. The results show that the mobile robots are able to travel clearly and completely safe with short path, proving the effectiveness of this method.  
Real time modified programmable universal machine for assembly (PUMA) 560 with intelligent controller Yousif I. Al Mashhadany; Wesam M. Jasim
Indonesian Journal of Electrical Engineering and Computer Science Vol 20, No 3: December 2020
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v20.i3.pp1194-1202

Abstract

In this work, the design of an integrated industrial application for use on a modified PUMA 560 robot arm was presented. The modified PUMA 560 robot has three joints; two of them are free-moving and the third one is at constant 90 degree angle. It has three links and two extra Griper links. Each joint was controlled via a DC motor through a PIC microcontroller. The design and implementation of modified PUMA 560 with electronic circuits to derive the motor were used with the robot and the working platform. These electronic circuits were also used to interface with the computer to control the DC motor based on the computer orders. The control signals used to control the application control system and to perform the defined tasks were received from a remote computer connected via internet. This design has been implemented in two phases; the first phase was the simulation of the complete control system, while, the second phase was the practical implementation. The obtained results were ensured the ability of the proposed system to perform the tasks of many industrial applications.
Hybrid approach for multi-objective optimization path planning with moving target Baraa M. Abed; Wesam M. Jasim
Indonesian Journal of Electrical Engineering and Computer Science Vol 29, No 1: January 2023
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v29.i1.pp348-357

Abstract

Path planning algorithms are the most significant area in the robotics field. Path planning (PP) can be defined as the process of determining the most appropriate navigation path before a mobile robot moves. Path planning optimization refers to finding the optimal or near-optimal path. Multi-objective optimization (MOO) is concerned with finding the best solution values that satisfy multiple objectives, such as shortness, smoothness, and safety. MOO present the challenge of making decisions while balancing these contradictory issues through compromise (tradeoff). As a result, there is no single solution appropriate for all purposes in MOO, but rather a range of solutions. Several objectives are considered as part of this study, including path security, length, and smoothness, when planning paths for autonomous mobile robots in a dynamic environment with a moving target. Particle swarm optimization (PSO) algorithms are combined with bat algorithms (BA) to make a balance between exploration and exploitation. PSO algorithms used to optimize two important parameters of the bat algorithm. The proposed solution is tested through several simulations based on varying scenarios. The results demonstrate that mobile robots can travel clearly and safely along short paths and smoothly, proving this method's efficiency.
A pre-trained model vs dedicated convolution neural networks for emotion recognition Asmaa Yaseen Nawaf; Wesam M. Jasim
International Journal of Electrical and Computer Engineering (IJECE) Vol 13, No 1: February 2023
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v13i1.pp1123-1133

Abstract

Facial expression recognition (FER) is one of the most important methods influencing human-machine interaction (HMI). In this paper, a comparison was made between two models, a model that was built from scratch and trained on FER dataset only, and a model previously trained on a data set containing various images, which is the VGG16 model, then the model was reset and trained using FER dataset. The FER+ data set was augmented to be used in training phases using the two proposed models. The models will be evaluated (extra validation) by using images from the internet in order to find the best model for identifying human emotions, where Dlib detector and OpenCV libraries are used for face detection. The results showed that the proposed emotion recognition convolutional neural networks (ERCNN) model dedicated to identifying human emotions significantly outperformed the pre-trained model in terms of accuracy, speed, and performance, which was 87.133% in the public test and 82.648% in the private test. While it was 71.685% in the public test and 67.338% in the private test using the proposed VGG16 pre-trained model.