Abstract: To ensure the accuracy of line of sight stabilization and improve the dynamic response capability, the conversely variable universe Fuzzy(CVU-Fuzzy)control method is used. This method changes the universe by operating with the input and output of the controller while keeping the quantization factor, scale factor and the actual universe constant. A contraction-expansion factor within 1 with simple form is adopted to set the Fuzzy reasoning system thus changing the universe conversely. Then optimise the parameters of contraction-expansion factors off-line by utilizing particle swarm optimization algorithm. A hybrid controller combines with CVU-Fuzzy and Pid is proposed to eliminate the static error. simulation results indicates that the method is feasible and effective.