Projects

This project focuses on predictive models from multi-sources of information to analyze the behavior and emotional state of the human for predicting future events and/or actions. The architecture can either be used to increase access to mental health services or to provide a better robot companion for the elderly people. The study is focusing on analysis and modelling using data collected from different smart phone apps, and sensors.The ultimate goal is to work with patient data (collected by clinical partners in the INTROMAT project) and elderly people data (collected by MECS project).

This project has as general objective to contribute to increase the acceptance of mobile robotics technologies in the area of services, focusing on the installation of mobile robot teams in large indoor spaces frequented by people (e.g., large shopping malls, common areas of shopping centers, offices and services, museums, etc.) to carry out patrol and surveillance missions independently. This project involves a joint effort of A company and two entities of the scientific and technological Portuguese system, in order to bring robots closer to human beings,helping them in the monotonous or repetitive tasks associated with the supervision, monitoring and surveillance of infrastructures, framed in the concept of multi-robot patrol for monitoring and surveillance of buildings and facilities.

The primary objective of the thesis was to select the optimal combination of features from functional near-infrared spectroscopy (fNIRS) using genetic algorithm (GA) for brain-computer interfacing (BCI). The objective of GA was to increase classification accuracy by selecting appropriate sub-features from the combination of features.

In this study, we examined the effects of using different classification modalities for the classification of a two class functional near-infrared spectroscopy- (fNIRS-) based brain-computer interface (BCI) according to a mental arithmetic task and rest experimental paradigm. It was shown that ANN has the highest classification accuracies among the classification modalities used in this study for both 2- and 3-dimensional feature sets derived from the changes in HbO concentrations signals across seven subjects. The results of this study represent a significant step forward in the on-going improvement of the classification accuracies of f NIRS-based BCI systems.

• The project was funded by Higher Education Commission (HEC) of Pakistan (grant no. SRGP-726).

In this study we examined the effects of using different combinations of six commonly used features for classification of a two-class functional near-infrared spectroscopy (fNIRS)-based BCI based on mental arithmetic and rest tasks. It was shown that the combination of the peak and mean values of the changes in the concentrations of oxygenated hemoglobin (HbO) and de-oxygenated hemoglobin (HbR) yielded the best average LDA classification results for 2- as well as 3-feature sets across seven subjects.

• The project was funded by Higher Education Commission (HEC) of Pakistan (grant no. SRGP-726).

The objective was to develop a 3R serial manipulator for sorting the objects depending on their shape placed on a conveyor for industrial use. The system had a visual feedback which is tracked and recognized using Image based visual servoing technique.

Developed the software of TIC TAC TOE Game using Graphical User Interface(GUI) in MATLAB®, Implemented different Path efficient control algorithms such as Depth first, Breadth first and A*.

Designed and fabricated a line tracking 4 wheeled mobile robot using path planning approach to reach at the required destination. The combination of LEDs and photo-transistors used as a sensor for detecting the line.

Designed and implemented a maze solving and ball potting robot for National Engineering and Robotic Competition 2012, held at EME NUST. Task of the robot is to follow the wall using infrared proximity Sensor, pot 3 different balls in their specified boxes and reach at the end (exit point) by solving the whole maze. Different path planning approaches were considered to minimize the completion time for the tasks.