MEMS gyroscope’s error modeling based on wavelet neural network of genetic algorithms
Zhou Bai-ling · Journal of Chinese Inertial Technology · 2008
A Wavelet Neural Network improved by Genetic Algorithms(GA) was presented.The method employs wavelet neural network as main approaching tool,which optimize the parameters by GA.Since the Wavelet neural network combines the self-learning ability of neural network with the time-frequency localization of the wavelet analysis,while the Genetic Algorithms(GA) employs the characteristic of global optimum searching,the wavelet neural network improved by GA holds strong approaching ability,which overcome the disadvantage of traditional modeling method.Its application in modeling a MEMS gyro’s random drift shows that the model has an identification error within1.75%.This precision can reach the requirements in engineering.