Integrating visual guidance and feedback for an industrial robot

Abstract

Visual servoing is a proven technique for the feedback control and guidance of robotic systems. Magnitude and clarity of the information collected from the surroundings to formulate the decision control affect the convergence of such strategies. This paper presents a supervisory control in visual servoing guaranteeing successful servoing, utilising the interaction of two cameras. The control is switched from the master camera observing the workspace to the end effector camera viewing the target. Relative Gain Array (RGA) analysis of the interaction matrix provides the sufficient condition for switching. The control scheme is validated with simulation and experimentation studies. The real-time communication between an ABB robot and the MATLAB environment is established through File Transfer Protocol (FTP).