CONTROLAB: an integrated system for intelligent control of a robot arm

Abstract

CONTROLAB is an environment which integrates intelligent systems and control algorithms aiming at applications in the area of robotics. Within CONTROLAB, two neural network architectures based on the backpropagation and the recursive models are proposed for the implementation of a robust speaker-independent word recognition system. The robustness of the system using the backpropagation network has been largely verified through use by children and adults in a totally uncontrolled environment such as large public halls for the exhibition of new technology products. Initial experimental results with the recursive network show that it will be able to overcome the backpropagation network major drawback, which is the frequent generation of false alarms. In addition, within CONTROLAB, a multivariable self-tuning algorithm is proposed to perform high-precision control of the trajectory, to be followed by a robot arm. High-precision control is achieved due to the ability of the algorithm to properly model the interaction forces among the robot arm joints and to cope with the non-linearities of the arm dynamics. CONTROLAB also includes computer vision facilities for object recognition and is currently running an application in which a SCARA robot arm is commanded by voice to pick up an object placed within its view area.