In order to reduce the influence of INS on the noise of laser gyro output

a method of adaptive network-based fuzzy inference system（ANFIS）with self-study ability is adopted to reduce the noise.A group of laser random data is used to model and validate the new method with simulations.And the power spectrum

mean value as well as the standard deviation are contrasted.The change of random error coefficient with Allan variance method is analysed.The result indicates that this method compensates effectively the laser gyro random error

which is propitious to enhance the precision of laser gyro outputs.