Mohd Zaidi Mohd Tumari

Feature Selection and Classifier Parameters Estimation for EEG Signals Peak Detection Using Particle Swarm Optimization

Electroencephalogram (EEG) signal peak detection is widely used in clinical applications. The peak point can be detected using several approaches, including time, frequency, time-frequency, and nonlinear domains depending on various peak features from several models. However, there is no study that provides the importance of every peak feature in contributing to a good and generalized model. In this study, feature selection and classifier parameters estimation based on particle swarm optimization (PSO) are proposed as a framework for peak detection on EEG signals in time domain analysis. Two versions of PSO are used in the study: (1) standard PSO and (2) random asynchronous particle swarm optimization (RA-PSO). The proposed framework tries to find the best combination of all the available features that offers good peak detection and a high classification rate from the results in the conducted experiments. The evaluation results indicate that the accuracy of the peak detection can be improved up to 99.90% and 98.59% for training and testing, respectively, as compared to the framework without feature selection adaptation. Additionally, the proposed framework based on RA-PSO offers a better and reliable classification rate as compared to standard PSO as it produces low variance model.

Composite Fuzzy Logic Control Approach to a Flexible Joint Manipulator

The raised complicatedness of the dynamics of a robot manipulator considering joint elasticity makes conventional model-based control strategies complex and hard to synthesize. This paper presents investigations into the development of hybrid intelligent control schemes for the trajectory tracking and vibration control of a flexible joint manipulator. To study the effectiveness of the controllers, a collocated proportional-derivative (PD)-type Fuzzy Logic Controller (FLC) is first developed for the tip angular position control of a flexible joint manipulator. This is then extended to incorporate a non-collocated Fuzzy Logic Controller, a non-collocated proportional-integral-derivative (PID) and an input-shaping scheme for the vibration reduction of the flexible joint system. The positive zero-vibration-derivative-derivative (ZVDD) shaper is designed based on the properties of the system. The implementation results of the response of the flexible joint manipulator with the controllers are presented in time and frequency domains. The performances of the hybrid control schemes are examined in terms of input tracking capability, level of vibration reduction and time response specifications. Finally, a comparative assessment of the control techniques is presented and discussed.

H ∞ controller with graphical LMI region profile for gantry crane system

This paper presents investigations into the development of H∞ controller with pole clustering based on LMI techniques to control the payload positioning of INTECO 3D crane system with very minimal swing. The linear model of INTECO 3D crane system is obtained using the system identification process. Using LMI approach, the regional pole placement known as LMI region combined with design objective in H∞ controller guarantee a fast input tracking capability, precise payload positioning and very minimal sway motion. A graphical profile of the transient response of crane system with respect to pole placement is very useful in giving more flexibility to the researcher in choosing a specific LMI region. The results of the response with the controllers are presented in time domains. The performances of control schemes are examined in terms of level of input tracking capability, sway angle reduction and time response specification. Finally, the control techniques is discussed and presented.

Feature Selection and Classifier Parameter Estimation for EEG Signal Peak Detection Using Gravitational Search Algorithm

Peak detection is a significant step in analyzing the electroencephalography (EEG) signal because peaks may represent meaningful brain activities. Several approaches can be used for peak point detection such as time domain, frequency domain, time-frequency domain, and nonlinear approaches. The main intention of this study is to find the significant peak features in time domain approach and this can be done using feature selection methods such as gravitational search algorithm (GSA) and particle swarm optimization (PSO). This study focuses on using GSA method, a new computational intelligence algorithm. Moreover, a rule-based classifier is employed to distinguish a peak point based on the selected features. Using GSA, the parameter estimation of the classifier and the peak feature selection can be done simultaneously. Based on the experimental results, the significant peak features of the peak detection algorithm were obtained where the average test accuracy is 77.74%.

Synchronous vs Asynchronous Gravitational Search Algorithm

Gravitational search algorithm (GSA) is a new member of swarm intelligence algorithms. It stems from Newtonian law of gravity and mass interaction. Typically the agents in GSA are updated synchronously, where the whole population is updated together after every members performance is evaluated. However, asynchronous update of agent has been used by other optimization algorithms. Therefore the performance of asynchronous GSA (A-GSA) is studied in this work. An agent in A-GSA is updated immediately after its performance evaluation. Hence an agent in A-GSA is updated without using complete and updated information of its entire population. Asynchronous update is more attractive from the perspective of parallelization. The results show that improvement to the straight forward implementation of A-GSA is needed.

Performance comparison between Sliding Mode Controller SMC and Proportional-Integral-Derivative PID controller for a highly nonlinear two-wheeled balancing robot

The research on two-wheels balancing robot has gained momentum due to their functionality and reliability when completing certain tasks. This paper presents investigations into the performance comparison of Sliding Mode Controller (SMC) and Proportional-Integral-Derivative (PID) controller for a highly nonlinear 2-wheels balancing robot. The mathematical model of 2-wheels balancing robot that is highly nonlinear is derived. The final model is then represented in state-space form and the system suffers from mismatched condition. Two system responses namely the robot position and robot angular position are obtained. The performances of the SMC and PID controllers are examined in terms of input tracking and disturbances rejection capability. Simulation results of the responses of the nonlinear 2-wheels balancing robot are presented in time domain. A comparative assessment of both control schemes to the system performance is analyzed and discussed.

Data-driven PID tuning based on safe experimentation dynamics for control of liquid slosh

An introductory research about a data-driven PID tuning for the control of liquid slosh system based on Safe Experimentation Dynamics (SED) is presented in this paper. A performance comparison between the SED and Simultaneous Perturbation Stochastic Approximation (SPSA) based method for data-driven PID tuning is observed and discussed. The performance is evaluated by numerical examples in terms of tracking performance, control input energy and computation time. The simulation results demonstrated that SED based datadriven PID successfully reduced the liquid slosh whilst the desired position of the cart is achieved. In addition, smaller control input energy is used.

Real-Time Nonlinear Complementary Filter on SO(3) for Attitude Estimation of Small-Scale Aerial Robot

This paper presents the real-time implementation of a powerful nonlinear complementary filter on special orthogonal group of rotation matrices, called as NCF SO(3) for attitude estimation. It fuses the raw data from accelerometers, magnetometer, and gyroscopes sensors to get reliable real-time attitude estimation. Gyroscopes is used as the main sensor for attitude estimation and another two sensors are used to correct drift error of gyroscopes. In this paper, the performance of NCF SO(3) is explored on performance in highly dynamic manoeuvres in real-time. Real-time experiments were conducted to compare its performance with conventional Extended Kalman Filter (EKF) to exploit the positive features of NCF SO(3) for small-scale aerial robot with limited on-board processor memory cases. The experimental results show the proposed real-time filter has excellent estimated attitude data and can reduce the computational cost, compared to EKF. Thus, it is suitable for small-scale aerial robot which has memory limitation of on-board processor.

Evaluation of Vector Evaluated Particle Swarm Optimisation Enhanced with Non-dominated Solutions and Multiple Nondominated Leaders Based on WFG Test Functions

Multi Objective Optimisation (MOO) problem involves simultaneous minimization or maximization of many objective functions. One of MOO algorithms is Vector Evaluated Particle Swarm Optimization (VEPSO) algorithm. In VEPSO, each objective function is optimised by a swarm of particles under guidance of the best solution, known as leader, from another swarm. Recently, an improved VEPSO algorithm, namely VEPSO incorporated non-dominated solution (VEPSOnds), has been introduced by the use of non-dominated solution as leader. Then, the VEPSOnds algorithm is further modified with multi leaders, namely VEPSO with multi leaders (VEPSOml). The improved VEPSO algorithms have been subjected to a series of numerical experiments based on ZDT benchmark datasets. In this study, a more complex benchmark datasets called WFG, is considered for the evaluation of VEPSO, VEPSOnds, and VEPSOml algorithms.

Development of straight line robot movement using Digital Compas Module

This project focuses on the development of straight line movement for a four-wheeled mobile robot. In this project, a DC gear motor is chosen as motion control for two driving wheels and the direction of robot will be controlled by servo motor to two steering wheels. PIC is selected as the brain board controller due to react and respond to the data received from Digital Compass Module to identify and figure out desired position. The implementation of internal PID algorithm is essentially used to restore the system to desired set-point position. Application of Digital Compass Module with the aid of PID control algorithm may command to drive the servo motor to go towards in straight line platform in accordance to the desired set-point direction has been fixed. The robotic hardware has been developed and analyzed successfully. As a result, in despite of unexpected external force varying the desired direction of robot, the robot would still be able to veer back to the original set-point direction to achieve a smooth and stabilized straight line movement.