Visual servo control of robotic manipulators based on artificial neural network

The authors describe a control scheme for a robotic manipulator system which uses visual information to position and orient the end effector. The main feature of the control scheme is direct integration of visual data into the serving process without any geometric process of the position and orientation, and without inverse kinematic calculation. Another feature is the use of an artificial neural network for the determination of the change in the joint angles required to achieve the desired position and orientation. This approach is effective because it essentially decomposes complex geometric calculations into a simple mapping of the network. The validity and the effectiveness of the proposed control scheme are verified by computer simulations