Authors:
B. Fevery
1
;
B. Wyns
1
;
L. Boullart
1
;
J. R. Llata García
2
and
C. Torre Ferrero
2
Affiliations:
1
Ghent University, Belgium
;
2
University of Cantabria, Spain
Keyword(s):
Active security, application, artificial vision, fuzzy logic, real-time, robot control.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Computational Intelligence
;
Control and Supervision Systems
;
Fuzzy Control
;
Fuzzy Systems
;
Human Augmentation and Shared Control
;
Image Processing
;
Industrial Networks and Automation
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Reasoning about Action for Intelligent Robots
;
Robotics and Automation
;
Soft Computing
;
Surveillance
;
Vision, Recognition and Reconstruction
Abstract:
An active security system assures that interacting robots don’t collide or that a robot operating independently doesn’t hit any obstacle that is encountered in the robots workspace. In this paper, an active security system for a FANUC industrial robot is introduced. The active security problem where one robot needs to avoid a moving obstacle in its workspace is considered. An obstacle detection and localization mechanism based on stereoscopic vision methods was successfully developed. To connect the vision system, an operator’s pc and the robot environment a real-time communication is set up over Ethernet using socket messaging. We used fuzzy logic for intelligent trajectory planning. A multitask oriented robot application in the KAREL programming language of FANUC Robotics was implemented and tested.