Fuzzy control of a class of hydraulically actuated industrial robots

Application of a fuzzy logic controller to a class of hydraulically actuated industrial robots is investigated in this paper, A simple, yet effective, set of membership functions and rules are described which meets certain control requirements of such robots, An off-line routine based on the simplex method is then outlined to tune the controller gains for an optimum response, The fuzzy control gains are tuned by minimizing the summation of absolute position errors over step input responses. The fuzzy logic controller is first examined through simulation of a two-link hydraulic robot, The robot has the same hydraulic configuration as many existing industrial manipulators, It is shown that the controller exhibits positive aspects which cannot be easily achieved by conventional control techniques, such as a PD controller with fixed gains, These aspects include a fast rise-time and a well maintained damped response, The fuzzy controller is then applied to an instrumented Unimate MK-II industrial hydraulic robot, The experimental results are encouraging in that the best performing control gains are found for different links with a reasonable number of trials and produce step responses with fast response and few oscillations at the set point. The controller demonstrates robustness in the presence of load, variation of set points, and nonideal hydraulic actuation systems. Compared to the simulation studies, the application of fuzzy control in experiments requires a reduction in the width of the zero output membership function to compensate for the effect of friction, leakage, and other nonlinearities inherent in the actual robot, The elimination of steady-state errors which were observed in both step and ramp input experimental responses is also addressed in this paper.