Authors:
Eugenio Aguirre
1
;
Miguel García-Silvente
1
;
Rui Paúl
1
and
Rafael Muñoz-Salinas
2
Affiliations:
1
University of Granada, Spain
;
2
University of Cordoba, Spain
Keyword(s):
Human-Robot Interaction, Interest Detection, Head Pose Estimation, Fuzzy Logic, Support Vector Machine.
Related
Ontology
Subjects/Areas/Topics:
Human-Robots Interfaces
;
Informatics in Control, Automation and Robotics
;
Robotics and Automation
Abstract:
Despite of the advances achieved in the past years in order to design more natural interfaces between intelligent systems and humans, there is still a great effort to be done. Considering a robot as an intelligent system, determining the interest of the surrounding people in interacting with it is an interesting ability to achieve. That information can be used to establish a more natural communication with humans as well as to design more sophisticated policies for resource assignment. This paper proposes a fuzzy system that establishes a level of possibility about the degree of interest that people around the robot have in interacting with it. First, a method to detect and track persons using stereo vision is briefly explained. Once the visible people is spotted, their interest in interacting with the robot is computed by analyzing its position and its level of attention towards the robot. These pieces of information are combined using fuzzy logic. The level of attention of a person
is calculated by analyzing the pose of his head that is estimated in real-time by a view based approach using Support Vector Machines (SVM). Although the proposed system is based only on visual information, its modularity and the use of fuzzy logic make it easier to incorporate in the future other sources of information to estimate with higher precision the interest of people. At the end of the paper, some experiments are shown that validate the proposal and future work is addressed.
(More)