Welcome to My Lab

Human-robot Symboitic Research

A knowledge based information modeling is being designed for autonomous humanoid service robots. The main objective is to implement a system of intelligent robots capable of interacting with human beings using natural communication to establish a symbiotic community supporting man-machine collaborative operations in a distributed working environment.

Al-Amin with the entertainment robot 'Aibo'.

When robots are working cooperatively with human-beings, it is necessary to share and exchange their ideas and thoughts. Gaze direction is an important non-verbal information that indicates the users intentions and interests. Eye tracking and gaze direction is, therefore, immerging tremendous interest in the advancement of human-robot interface since it provides a natural and efficient way of exploring expressions for the interaction.      A real-time eye tracking and gaze direction system has already been implemented for human-robot interaction. Based on the position and movement of the eyes, the system determines where on the computer's screen the user is looking. The user can make a selection by moving his/her eyes in different directions and depending on this selection, the system provides instruction for controlling a robot, named AIBO using his/her gaze direction. The system is based on visual and geometrical information of the user's face from the video streams and is organized with the detection of face depending on the similarity measure of the hue components of the images in the HSV color histograms. Eyes are then extracted from face skeleton with the knowledge of the face geometry. Eye tracking is established by the computation of the optical flow in consecutive frames of the video sequences.      The range of the robot's interaction with human beings is limited by many factors, including its visual perception. As new percepts are added, new dimensions of behavior are possible to include, for example face identification, facial expression, lip movement, head orientation, etc. Our next approach is to make the system capable of detecting facial gestures and expressions and include more instructions for different robots, such as AIBO, ROBOVIE, SCOUT, PINO, HOAP, and so on.

LINKS:

IEEE transaction on PAMI

National Institute of Informatics

Sciencedirect Journals

IEEE Digital Library

Computer Vision and Image Understanding

Graphical Models and Image Processing

ACM Digital Library

Some Publications:

1. Identification of Actors drawn in Ukiyoe pictures, Pattern Recognitoin, Vol. 35, No. 1, 2002, pp. 93-102.

2. Recovering the control points of Bezier curves for line image indexing, Electronic Imaging, Vol. 11. No. 2, 2002, pp. 177-186.

3. A fast algorithm for finding control points of Bezier curves, International Journal of Robotics and Automation, Vol. 16. No. 3, 2001. pp. 117-123.