Based on adaptive neuro-fuzzy inference system(ANFIS), corrections of position and velocity are provided for
inertial navigation system(INS) when global position system(GPS) signal is blocked to improve the accuracy and robustness
of the whole system. To achieve the object, wavelet is used to de-noise the output of the sensors, then positions and velocities
are set to be the input of ANFIS, while position and velocity corrections are output after trained. The desired outputs of
trained samples are the position and velocity error after wave multi-resolution analysis. The result shows that, when GPS is
blocked, position accuracy is increased forty percent than Kalman filtering, and this method can provide reliable and effective
positioning services for vehicles.