Semi-autonomous robotic manipulation

Abstract

Semi-autonomous intelligent robotic systems measure their environment using a variety of sensors. The design and implementation of such a robotic system that is equipped with multiple sensors is presented. The example task involves the insertion of a printed circuit card into a card cage. Intelligent behavior is accomplished by coupling the task execution sequence with information derived from three different sensors: a fixed overhead three-dimensional vision system, a fingertip infrared beam sensor, and a six degree-of-freedom wrist-mounted force-torque sensor. The vision system is used to locate the printed circuit cards and the insertion slots. Information provided by the fingertip sensor is used to complement the information from the vision system to obtain a proper grasping location on the printed circuit card. A fuzzy logic-based system is used to interpret the forces and torques measured by the force-torque sensor during the insertion process. The fuzzy controller accommodates uncertainties in the workspace environment. The robotic system is able to accomplish the example task without full time human supervision.