Aiming at the problem that MEMS gyroscope bias is greatly affected by temperature, a gyro bias temperature compensation method based on particle swarm optimization (PSO) algorithm and radial basis function (RBF) neural network is proposed. Through the analysis and preprocessing of the temperature error, the particle swarm optimization algorithm is used to search for the optimal configuration of the RBF neural network, and then the RBF neural network coefficients are further optimized. The experimental results show that in the temperature range of -40℃ to +60℃, the temperature compensation method of PSO-RBF neural network can reduce the maximum error and standard deviation of the MEMS gyroscope''s zero bias to 0.0034°/s and 0.0013°/s, the feasibility of the method is verified, and it has strong engineering application value for improving the accuracy of gyroscope measurement.