Volume 15, Issue 1

Independent Component Analysis (ICA) has been successfully applied to a variety of problems, from speaker identification and image processing to functional magnetic resonance imaging (fMRI) of the brain. In particular, it has been applied to analyze EEG data in order to estimate the sources form the measurements.
However, it soon became clear that for EEG signals the solutions found by ICA often depends on the particular ICA algorithm, and that the solutions may not always have a physiologically plausible interpretation. Therefore, nowadays many researchers are using ICA largely for artifact detection and removal from EEG, but not for the actual analysis of signals from cortical sources. However, a recent modification of an ICA algorithm has been applied successfully to EEG signals from the resting state. The key idea was to perform a particular preprocessing and then apply a complexvalued
ICA algorithm.
In this paper, we consider multiple complex-valued ICA algorithms and compare their performance on real-world resting state EEG data. Such a comparison is problematic because the way of mixing the original sources (the “ground truth”) is not known. We address this by developing proper measures to compare the results from multiple algorithms. The comparisons consider the ability of an algorithm to find interesting independent sources, i.e. those related to brain activity and not to artifact activity. The performance of locating a dipole for each separated independent component is considered in the comparison as well.
Our results suggest that when using complex-valued ICA algorithms on preprocessed signals the resting state EEG activity can be analyzed in terms of physiological properties. This reestablishes the suitability of ICA for EEG analysis beyond the detection and removal of artifacts with real-valued ICA applied to the signals in the time-domain.

Training the user in Brain-Computer Interface (BCI) systems based on brain signals that recorded using Electroencephalography Motor Imagery (EEG-MI) signal is a time-consuming process and causes tiredness to the trained subject, so transfer learning (subject to subject or session to session) is very useful methods of training that will decrease the number of recorded training trials for the target subject. To record the brain signals, channels or electrodes are used. Increasing channels could increase the classification accuracy but this solution costs a lot of money and there are no guarantees of high classification accuracy. This paper introduces a transfer learning method using only two channels and a few training trials for both feature extraction and classifier training. Our results show that the proposed method Independent Component Analysis with Regularized Common Spatial Pattern (ICA-RCSP) will produce about 70% accuracy for the session to session transfer learning using few training trails. When the proposed method used for transfer subject to subject the accuracy was lower than that for session to session but it still better than other methods.

The identification of system parameters plays an essential role in system modeling and control. This paper presents a parameter estimation for a permanent magnetic DC motor using the simulink design optimization method. The parameter estimation may be represented as an optimization problem. Firstly, the initial values of the DC motor parameters are extracted using the dynamic model through measuring the values of voltage, current, and speed of the motor. Then, these values are used as an initial value for simulink design optimization. The experimentally inputoutput data can be collected using a suggested microcontroller based circuit that will be used later for estimating the DC motor parameters by building a simulink model. Two optimization algorithms are used, the pattern search and the nonlinear least square. The results show that the nonlinear least square algorithm gives a more accurate result that almost approaches to the actual measured speed response of the motor.

In this paper, the hybrid-climbing legged robot is designed, implemented, and practically tested. The robot has four legs arranged symmetrically around the body were designed for climbing wire mesh fence. Each leg in robot has 3DOF which makes the motion of the robot is flexible. The robot can climb the walls vertically by using a unique design of gripper device included metal hooks. The mechanism of the movement is a combination of two techniques, the first is the common way for the successive movement like gecko by using four limbs, and the second depending on the method that used by cats for climbing on the trees using claws, for this reason, the robot is named hybrid-climbing legged robot. The movement mechanism of the climbing robot is achieved by emulating the motion behavior of the gecko, which helped to derive the kinematic equations of the robot. The robot was practically implemented by using a microcontroller for the mainboard controller while the structure of the robot body is designed by AutoCAD software. Several experiments performed in order to test the success of climbing on the vertical wire mesh fence.

In maze maneuvering, it is needed for a mobile robot to feasibly plan the shortest path from its initial posture to the desired destination in a given environment. To achieve that, the mobile robot is combined with multiple distance sensors to assist the navigation while avoiding obstructing obstacles and following the shortest path toward the target. Additionally, a vision sensor is used to detect and track colored objects. A new algorithm is proposed based on different type of utilized sensors to aid the maneuvering of differential drive mobile robot in an unknown environment. In the proposed algorithm, the robot has the ability to traverse surrounding hindrances and seek for a particular object based on its color. Six infrared sensors are used to detect any located obstacles and one color detection sensor is used to locate the colored object. The Mobile Robotics Simulation Toolbox in Matlab is used to test the proposed algorithm. Three different scenarios are studied to prove the efficiency of the proposed algorithm. The simulation results demonstrate that the mobile robot has successfully accomplished the tracking and locating of a colored object without collision with hurdles.

In this paper, a new algorithm called table-based matching for multi-robot (node) that used for localization and orientation are suggested. The environment is provided with two distance sensors fixed on two beacons at the bottom corners of the frame. These beacons have the ability to scan the environment and estimate the location and orientation of the visible nodes and save the result in matrices which are used later to construct a visible node table. This table is used for matching with visible-robot table which is constructed from the result of each robot scanning to its neighbors with a distance sensor that rotates at 360°; at this point, the location and identity of all visible nodes are known. The localization and orientation of invisible robots rely on the matching of other tables obtained from the information of visible robots. Several simulations implementation are experienced on a different number of nodes to submit the performance of this introduced algorithm.

Optical OFDM based on discrete Hartley transform (DHT-O-OFDM) has been proposed for large-size data mapping intensity modulation direct detection (IM/DD) scheme as an alternative to the conventional optical OFDM. This paper presents a performance analysis and evaluation of IM/DD optical DC-biased DHT-O-OFDM over diffused multipath optical wireless channels. Zero-padding guard interval along with minimum mean-square error (MMSE) equalizer are used in electrical domain after the direct detection to remove the intersymbol interference (ISI) and eliminate the deleterious effects of the multipath channels. Simulation results show that the ZP-MMSE can effectively reduce the effects of multipath channels. The results also show that the effects of optical wireless multipath channel become more serious as the data signaling order increases.

In coordination of a group of mobile robots in a real environment, the formation is an important task. Multimobile robot formations in global knowledge environments are achieved using small robots with small hardware capabilities. To perform formation, localization, orientation, path planning and obstacle and collision avoidance should be accomplished. Finally, several static and dynamic strategies for polygon shape formation are implemented. For these formations minimizing the energy spent by the robots or the time for achieving the task, have been investigated. These strategies have better efficiency in completing the formation, since they use the cluster matching algorithm instead of the triangulation algorithm.