The publication is a description of an adaptive and auto-instructive adjustment program in fuzzy technology of the single-arm movement of a robot arm. The program was developed and simulated in the Matlab environment, using fuzzy technology-specific functions, superior to other adaptive and non-interactive programming techniques. The Fuzzy Model Reference Learning Control (FMRLC) designed to command an arm of a robot, is much easier to parameterize than any adaptive adjustment scheme with a robust local standard model, for example Lyapunov type. This is because that regulator admits many initial variants, which will later be modified by the adaptation block, which acts on the rule base.