Abstract：A learning control method for an robot manipulator is introduced.The acceleration error is used to correctthe motion of the industrial manipulator motors.A theoretical method which estimates the convergence con-dition of the learning control process based on the limit condition of a geometric series is proposed.The valid-ity of the proposed theory is proved through a computer simulation of PUMA-562 robot.
 Arimoto S,Kawamura S,Miyazaki F. Bettering operation of robots by learning. J Robotic Systems, 1984;1(2):123-140.
 Arimoto S,ICawamura S,Miyazaki F. Tamaki S. Learning control theory for dynamical systems. In, Proc 24th IEEE CDC, Florida, USA (Dec 1985)
 Neuman C P N, Murray J J,The complete dynamic model and customized algorithms of the PUMA robot. IEEE Trans Systems, Cybernetics. 19871 SMC-17; 635-643