Fuzzy SOC (Self-Organising Control) algorithm is a control law based on rules that governs the nature of controlled plant and is an AI control method that performs automatic creation and modification of the rules. The objective of this paper is to apply Fuzzy SOC algorithm to axis control of multi-axis robot and to investigate the performance of the Fuzzy SOC algorithm through robot axis control experiment. The PUMA 760 robot manipulator is used for an experimental test model. For the experiment servo axis control boards are newly designed using INTEL 8032, 8-bit micro controller for interfacing with a personal computer. The Fuzzy SOC algorithm is implemented to an IBM-AT/386 personal computer which communicates with the axis control boards through parallel I/O interfaces. In the experiment the first 3 joints of the robot manipulator are controlled simultaneously under various experimental conditions, such as various types of input trajectory, variation of control parameters, to investigate the performance and effect of the proposed control algorithm. From the experimental results, it can be concluded that Fuzzy SOC algorithm is effective for control of a system which is difficult to model and under complex control environments. Also, it can be concluded that the proposed algorithm is a computationally efficient scheme which requires small computation time and is applicable to the servo control of the fast dynamic system as robot manipulator.