Abstract: A methodology is presented whereby a neural network is used to learn the inverse kinematic relationship for the position and orientation of a manipulator.For the first time the arm solution for the orientation of a manipulator using a self-organizing neural net is studied in this paper.By thoroughly studing on Martinetz、Ritter and Schulten′s self organizing neural network based on Kohonen′s self-organizing mapping algorithm and Widrow Hoff type error correction rule and closely combining the characters of the inverse kinematic relationship for a robot arm,a new training model of the self-organizing neural network is proposed.The computer simulation results for a PUMA 560 robot show that this method gives great imporovement in self-organizing capability and precision in training process.