GPS uses the radio signals transmitted by satellites to determine the position of the user receiver on the earth bycalculating the relative position between the satellite and the GPS receiver, and can provide real-time positioning, speed measurementand timing services for the user receiver. The traditional receiver uses the least squares method to solve the positioning, and thecoordinates of the receiver are prone to large jumps. By using the principle of Kalman filtering, a model is established according tothe motion characteristics of the receiver, and the relationship between the results of each navigation solution is used to smooth thepositioning results and eliminate the positioning jitter error caused by the relative independence of the results of the least squaresalgorithm. The positioning test results show that the positioning algorithm designed in the paper has a good positioning effect.