Visual based control system of a 3-CPU translational parallel manipulator

Recent advances in vision devices and image processing has allowed to integrate vision sensors into the control architecture of robot systems to include information regarding robot’s environment: in this case the feedback loop handles operational space variables based on a real time estimation of the end-effector pose obtained by means of an elaboration of visual measurements. The paper presents an application of visual servoing on a 3-CPU translational parallel manipulator with a camera mounted directly on the moving platform, in the so-called eye-in-hand configuration. After an introductory description of the robot and of the control and vision hardware, the calibration process of the camera is explained and a PD controller with gravity compensation is presented.