A Force/Positiun Hybrad Self-tuning Cuntrol Of Manipulaturs

A method of implicit self-tuning control of manipulators with feedforward compensation of normal torques is presented in this paper. The method is based on perturbed model and weighted minimum variance control theory. The position and force errors expressed in taks-space coordinates are used as control signals. Each degree of freedom in task, frame are controlled by either position or force. The modelling and control design are implemented for each degree of freedom separately. The PUMA 560 manipulator has been used as the model for the study of the force/position hybrid seif-tuning control. The simulations implemented on Dual 68000 computer, shows the effectiveness of the method. Finally the computation complex is analysed and shows a good prospect for practical use of the method.