Integration of vision systems to robotic cells enlarges the potential application of manufacturing systems. Related to this integration is the calibration of cameras which needs the knowledge of the robot and cameras geometries. In this paper we exploit the approximation capabilities of neural networks to avoid the computation of the robot inverse kinematics as well as the inverse task space-camera mapping which involves tedious calibration procedures. The feasibility of the proposed neural controller is illustrated through experiments on a planar robot.
展开▼