Abstract:
Vision based robot applications have taken a great deal of attention, with the development of electronic and computer technology. The visual feedback loop is very effective for improving the dexterity and flexibility. In this study, application of real time visual servoing approach is presented that enables a robot to robustly execute arm grasping and manipulation tasks. This task is decomposed on four stages a) finding object b) determining object's pose c) moving the robotic arm from an initial position towards the object d) grasping the object. The robot used in this work consists of an arm and head parts. The robotic arm has six degree of freedom, five degree of freedom are located at the arm while one degree of freedom is assigned to the gripper. Head has two degree of fredom which is pan-tilt platform. The image-based control strategy is designed using Fuzzy-PID controller. In this way, position error between target object and griper is minimized and the gripper can grasp the target object precisely. Real-time implementation of the proposed method is carried out using Matlab-Simulink. Experimental results show that, the developed design is valid to detect and grasp objects in real time.
