A fuzzy control algorithm reducing steady-state position errors of robotic manipulators

Abstract

Conventional fuzzy controllers for the motion tracking of robotic manipulators utilize two input variables, position errors and velocity errors, to deal with highly nonlinear and time-varying dynamics associated with robot motion, and position errors usually remain at steady state conditions. In order to eliminate the steady state position errors, a new fuzzy control algorithm is proposed using three input variables of the fuzzy controller: position error, velocity error, and integral of position errors. Simulation and experiment studies show that the position errors at steady state are decreased more than 90% compared to those of conventional fuzzy control when the proposed fuzzy control is applied to the 2-axis direct-drive SCARA robot.