nan Suprijanto

Design of Brain-computer interface platform for semi real-time commanding electrical wheelchair simulator movement

Brain Computer Interface (BCI) research has developed intensively during past years, especially in the assistive technology development for disabled people, e.g. electric wheelchair. This paper will propose our design of the BCI platform for commanding the electric wheelchair simulator. A wireless EEG device is used because of the need for mobile, comfortable, and user-friendly BCI platform. This paper will cover the design process to build the BCI platform prototype, and evaluation about the result achieved. Our BCI platform consists of Emotiv EPOC wireless EEG, and a computer for EEG acquisition, noise filtering, feature extraction, feature classification, and gives control command to wheelchair simulator that is connected via USB. Power spectral density is used as feature extraction of μ rhythm and artificial neural network with multi layer perceptron is used for classifying the idle and movement condition and for classifying between right hand and left hand movement. The first prototype of our BCI system has an interchangeable classification database that allows drastic modification of the system input, output, user recognition, and even the classification method.

A collaborative control of brain computer interface and robotic wheelchair

This paper presents a new scheme in collaborating non-invasive brain computer interface (BCI) and a wheelchair equipped with robotic systems. The BCI system implements steady state visual evoked potential (SSVEP) method that extracts features from electro-encephalography (EEG) signals in determining the intentions or commands from human-brain. The classification of EEG signals utilizes filter-bank in accordance with frequency of the stimuli visual. The user intentions, which are to move “left”, “right”, and “forward”, and “stop” are collaborated with an intelligent robotic system of a wheelchair. The wheelchair system is equipped with environment recognition sensors. The collaborative control is to manage the motion of the wheelchair based upon the intention of the user and the condition of the environment. The scenario is limited to the collaboration of BCI-robotic wheelchair in a corridor environment with walls on both sides. The user intention is set to “forward”. The intelligent robotic wheelchair perform wall following and obstacle avoidance motions while receiving the command “forward”. Also, it performs emergency-stop if the extracted intentions from the BCI put the user in dangerous situation. Experimental results are conducted to show the effectiveness of the proposed method.

Towards online application of wireless EEG-based open platform Brain Computer Interface

Brain Computer Interface (BCI) is a system that directly utilize Electroencephalograph (EEG) signals to control external devices without aid from any limb of the body. BCI system consists of brainwave acquisition, signal processing, feature extraction and classification. A design of BCI system has been developed by using a wireless EEG Emotiv EPOC neuroheadset and OpenViBE. Both of them are open-source system which gives opportunity to develop our BCI system freely. Mu wave is extracted from the acquired brainwaves when the subject imagined hand movement. Mu wave can be obtained on FC5 and FC6, where premotor activities take place, by apply it to a 8 - 13 Hz bandpass filter. Mu wave power which is the square of EEG signal amplitude is extracted to be classified into two different classes. Feature classification is done by using Support Vector Machine (SVM) in offline classification and online training. EEG signal was acquired on three healthy subjects without well training with BCI control. The task of subjects are imaginary movement of right and left hand with stimulation by a left and right arrow on the screen. Configuration for training and testing phase has been successfully done in OpenViBE towards online application. The mean recognition rate in offline testing and single trial classification is 60.63% for right arrow and 45.93% for left arrow on all subjects.

Wall following control for the application of a brain-controlled wheelchair

This paper presents the implementation of a wall following control for the application of a brain-controlled wheelchair. The control objective is to make electric wheelchair can go forward at a certain distance and at zero angle with a wall. The electric wheelchair is equipped with ultrasonic transducers to detect the distance and the angle between the wheelchair and the wall. The control algorithm is derived using Lyapunov method and the stability of the system is assured. The wheelchair is also combined with adaptive control in order that the desired speed of both motors is achieved and to make the performance of wall following motion more stable. The effectiveness of the proposed method is shown by experimental results.

Quantitative image analysis of periapical dental radiography for dental condition diagnosis

Periapical dental radiography have been used by the dentist to diagnose the lessions of the tooth. In Indonesia, recent development of this dental technology still used conventional method by using negative film with several limitation. This research was conducted to study digitalization of periapical film using digital dental X Ray reader for dental condition diagnosis. This dental condition was made by two step image analysis, consist of qualitative analysis and quantitative analysis the dentin and pulp. To segmen dentin and pulp, active contour was used to simplify the image analysis process. Qualitative analysis was made by the help of the visual inspection of the dentist, while quantitative analysis was made by compute various statistic parameter. Result of this research show that varians and the intensity ratio between dentin and pulp is good enough to be statistic parameter to differentiate the condition of the dental.

Wall following control of a mobile robot without orientation sensor

The paper presents a novel wall following control of a mobile robot without orientation sensors. The problem is formulated as a path following problem of a mobile robot. The control law derived using the proposed method assure the asymptotic stabilization of the origin by using the Lyapunov method. The effectiveness of the proposed method is shown by experimental results using a P3DX mobile robot utilizing sonar sensors to detect only the distance to the wall without orientation sensor.

Mobile robot localization using modified particle filter

Localization is one of many issues in mobile robot study. Localization is an essential ability for mobile robot to determine its location, so that it can plan a movement and go to a desired location. The mutual method for mobile robot localization is using a particle filter. High computation needs in particle filter is one of problem in particle filtering to get accurate location. The paper proposes a low computational mobile robot localization using a particle filter. It uses two methods: local localization using a dead reckoning method and global localization using a landmark-based vision sensor. Simulation results show that the proposed method provides a good estimation on the mobile robot position and orientation.

P300 detection using nonlinear independent component analysis

In this paper, a nonlinear independent component analysis (NICA) extraction method for brain signal based EEG-P300 are proposed. The performance of the proposed method is investigated through a comparison of well-known extraction methods (i.e., AAR, JADE, and SOBI algorithms). Finally, the promising results reported here reflect the considerable potential of EEG for the continuous classification of mental states.

Heuristic Steady State Visual Evoked Potential based Brain Computer Interface system for robotic wheelchair application

Development of Brain Computer Interface (BCI) system to enable connection of human intention directly with automated appliances without going through peripheral muscle these days is increasing as an effort to provide solution for patients with disabilities to do their activity normally. Unfortunately current BCI system still has shortage in complex configuration during electroencephalograph (EEG) measurement that uses large numbers of electrodes causing difficulties for the application of the BCI system especially for common usage. Therefore this study, we conduct research on Steady State Visual Evoked Potential (SSVEP)-based Brain Computer Interface to design BCI system utilizing minimum amount of electrodes which is able to use for robotic wheelchair movement control. The BCI system is design heuristically based from the collected experiment data, utilizing banks of filter for the feature extraction and threshold-voting system for feature classification. Offline evaluation of the designed BCI system shows the average correct classification is 84,94% with Information Transfer Rate (ITR) is 68,9440 bits/min.

Microscopic Surface Measurement Using Phase Shifting Method

Accurate information of microscopic topography is very important for efficacy assessment of a surface texture of skin health. Due to the limitations of the direct visual assessment of skin microscopic topography, an optical dermastocopy is very common to be used as skin imaging device to magnify skin topography based on a white light reflection. The limitation of this method is its poor spatial resolution to quantify skin topography. In this work, microscopic skin imaging based on phase shifting method is configured using a DLP pico-projector with LED illumination and a handheld digital microscope. As illuminator for the digital microscope, the DLP projector is programmed to generate patterned light on skin surface. Image processing is required in providing accurate information of surface topography. The first step, a wrapped phase shifting must be extracted from acquired intensity images. The second step is obtaining unwrapped phase image, which is a critical process because it must be recovered from wrapped phase shifting that containednoise. Finally, phase offset due to multiples of 2π during phase unwrapping must be removed. Early experiments on simple object are carried out to test the level of distortion of fringe in several variations of contrast and also to test the performance of the system on several frequency variations. The test results indicate the depth proportion obtained from absolute phase image has the same trend as the proportion of direct measurement. Implementation on the skin surface profile performed on three test areas: the back of the hand and knuckle creases. Based on quantitative and qualitative analysis,our proposed scheme of skin imaging based on phase shifting is promising for surface profile measurement and imaging of the skin.