Luhur Bayuaji
Universiti Malaysia Pahang

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

Parameter Prediction for Lorenz Attractor by using Deep Neural Network Nurnajmin Qasrina Ann; Dwi Pebrianti; Mohammad Fadhil Abas; Luhur Bayuaji; Mohammad Syafrullah
Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol 8, No 3: September 2020
Publisher : IAES Indonesian Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52549/ijeei.v8i3.1272

Abstract

Nowadays, most modern deep learning models are based on artificial neural networks. This research presents Deep Neural Network to learn the database, which consists of high precision, a strange Lorenz attractor. Lorenz system is one of the simple chaotic systems, which is a nonlinear and characterized by an unstable dynamic behavior. The research aims to predict the parameter of a strange Lorenz attractor either yes or not. The primary method implemented in this paper is the Deep Neural Network by using Phyton Keras library. For the neural network, the different number of hidden layers are used to compare the accuracy of the system prediction. A set of data is used as the input of the neural network, while for the output part, the accuracy of prediction data is expected. As a result, the accuracy of the testing result shows that 100% correct prediction can be achieved when using the training data. Meanwhile, only 60% correct prediction is achieved for the new random data.
PID Controller Design for Mobile Robot Using Bat Algorithm with Mutation (BAM) Dwi Pebranti; Luhur Bayuaji; Yogesvaran Arumgam; Indra Riyanto; Muhammad Syafrullah; Nurnajmin Qasrina Ann Ayop
Proceeding of the Electrical Engineering Computer Science and Informatics Vol 6: EECSI 2019
Publisher : IAES Indonesia Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/eecsi.v6.1996

Abstract

By definition, a mobile robot is a type of robotthat has capability to move in a certain kind of environmentand generally used to accomplish certain tasks with somedegrees of freedom (DoF). Applications of mobile robots coverboth industrial and domestic area. It may help to reduce risk tohuman being and to the environment. Mobile robot is expectedto operate safely where it must stay away from hazards such asobstacles. Therefore, a controller needs to be designed to makethe system robust and adaptive. In this study, PID controller ischosen to control a mobile robot. PID is considered as simpleyet powerful controller for many kind of applications. Indesigning PID, user needs to set appropriate controller gain toachieve a desired performance of the control system, in termsof time response and its steady state error. Here, anoptimization algorithm called Bat Algorithm with Mutation(BAM) is proposed to optimize the value of PID controller gainfor mobile robot. This algorithm is compared with a wellknownoptimization algorithm, Particle Swarm Optimization(PSO). The result shows that BAM has better performancecompared to PSO in term of overshoot percentage and steadystate error. BAM gives 2.29% of overshoot and 2.94% ofsteady state error. Meanwhile, PSO gives 3.07% of overshootand 3.72% of steady state error.
Intelligent Control for Visual Servoing System Dwi Pebrianti; Ong Ying Peh; Rosdiyana Samad; Mahfuzah Mustafa; N. R.H Abdullah; Luhur Bayuaji
Indonesian Journal of Electrical Engineering and Computer Science Vol 6, No 1: April 2017
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v6.i1.pp72-79

Abstract

This paper presents intelligent control for visual servoing system. The proposed system consists of a camera placed on a Pan Tilt Unit (PTU) which consists of two different servo motors. Camera and PTU are connected to a personal computer for the image processing and controlling purpose. Color threshold method is used for object tracking and recognition. Two different control methods, PID and Fuzzy Logic Control (FLC) are designed and the performances are compared through simulation. From the simulation result, the settling time of PID controller is 40 times faster than FLC. Additionally, the rise time of PID is about 20 times faster than FLC. However, the overshoot percentage of PID controller is 4 times higher than FLC. High overshoot value is not preferable in a control system, since it will cause the damage to the system. Real implementation of FLC on a home-built visual servoing system is conducted. Two different types of FLC, 9 and 11 rules of FLC are designed and implemented on the system. The experimental result shows that FLC with different total number of rules give different system performance. The settling time of FLC with 11 rules is 2 times faster than FLC with 9 rules. Additionally, the overshoot percentage of FLC with 11 rules is 2 times lower than FLC with 9 rules.
Automated-tuned hyper-parameter deep neural network by using arithmetic optimization algorithm for Lorenz chaotic system Nurnajmin Qasrina Ann; Dwi Pebrianti; Mohd Fadhil Abas; Luhur Bayuaji
International Journal of Electrical and Computer Engineering (IJECE) Vol 13, No 2: April 2023
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijece.v13i2.pp2167-2176

Abstract

Deep neural networks (DNNs) are very dependent on their parameterization and require experts to determine which method to implement and modify the hyper-parameters value. This study proposes an automated-tuned hyper-parameter for DNN using a metaheuristic optimization algorithm, arithmetic optimization algorithm (AOA). AOA makes use of the distribution properties of mathematics’ primary arithmetic operators, including multiplication, division, addition, and subtraction. AOA is mathematically modeled and implemented to optimize processes across a broad range of search spaces. The performance of AOA is evaluated against 29 benchmark functions, and several real-world engineering design problems are to demonstrate AOA’s applicability. The hyper-parameter tuning framework consists of a set of Lorenz chaotic system datasets, hybrid DNN architecture, and AOA that works automatically. As a result, AOA produced the highest accuracy in the test dataset with a combination of optimized hyper-parameters for DNN architecture. The boxplot analysis also produced the ten AOA particles that are the most accurately chosen. Hence, AOA with ten particles had the smallest size of boxplot for all hyper-parameters, which concluded the best solution. In particular, the result for the proposed system is outperformed compared to the architecture tested with particle swarm optimization.