The Study Of Uncertain Robot Force/Position Intelligent Control And Trajectory Tracking Experiment

Robot force/position control problem is the most interesting topic in the robot control. In this paper, parameter perturbation and the uncertainties of environment stiffness are mainly studied in the robot control. The main research of this paper is concluded as follows: Firstly, the three control schemas in force/position control are presented: 1. For uncertain robot this paper simplifies cost and structure by using PD+feedforword control structure based on robust adaptive control. The method identifies uncertain upper boundary function by using RBF NN, so the system has more adaptive ability. A neural network control method in the force loop is presented based on force/position hybrid control. This method can improve the adaptive ability of contact stiffness uncertainty and avoid solving regression matrix. 2. Because the FNN can make up for each other, this paper presents a control schema that FNN controller combines with proportion controller. Uncertain upper boundary function is learned on-line by using FNN, and proportion controller strengthens the completeness of FNN control strategy. FNN is trained by using disparity target learning error produced by FIE. This can avoid supersaturating problem made by feedback error directly, and restrain influence of measuring noise and improve control performance. 3. A neural network control design with mixed H₂ /H_∞ performance was proposed because of the different requirement to uncertainty, which is to respectively compensate for parameter uncertainties and external disturbances of robot system. This method can ensure the robust stability under a prescribed attenuation level for the external disturbance, and also can satisfy H₂ optimal performance index. NN learns parameter uncertainties of system when the system satisfies H₂ /H_∞ optimal performance. Secondly, the theory of hybrid force/position control is clear, however, it is difficult to implement in practice. Force control is combined with position control in the same structure by impedance control. They are controlled by the same method, thus mean less work. There are two main problems in robot impedance control. One is the uncertainty of robot dynamics and environment. This paper presents an impedance control strategy with robust performance and uses fuzzy neural network to compensate for the uncertainties in force control. The other is the tracking ability of desired force in impedance control. A fuzzy neural control scheme is presented based on position control and FNN controller learns the reference position. It is shown that the control schema can improve force control precision. Finally, a fuzzy adaptive PD control strategy is presented in trajectory of uncertain robot and force/position hybrid control. From the simulation result, it can be seen that the method can guarantee good real time and be easily realized. This paper carries out trajectory control experiment study based on fuzzy adaptive PD control strategy. The equipment SRV02 (2-DOF) Robot used by the experiment is made in Canada. The experiment result illustrates the proposed method can control the trajectory of robot well.