The direct adaptive trajectory control of robot manipulators via fuzzy logic algorithm

The paper is addressed to study the trajectory control of robotic mechanisms, particularly, to the transient performance of path tracking control systems. The coefficients and terms of the robot dynamic equations may be partly unknown and some dynamic effects may be unmodeled. One general scheme, which is a combination of the generalization of the computed torque method and the velocity gradient technique, provides a simple way to obtain some known and unknown parameter adaptive control laws, in particular, the exponential path tracking adaptive control algorithms with the important properties of robustness to bounded disturbances and unmodeled dynamics. The algorithms require the on-line measurements of the tracking error and its first derivative. The persistent excitation assumptions for desired paths are not required. The fuzzy logic control strategy is applied to obtain implement able algorithms on the base of the developed adaptive laws. The approach includes the attractive features of both strategies: robustness from direct adaptive control, and simple implementability from fuzzy logic control.