To improve the accuracy and precision of automatic navigation system for agricultural machinery

this paper designs an intelligent combined navigation system for agricultural machinery based on Global Positioning System (GPS) and Inertial Navigation System (INS) for the problems of relatively large measurement error of low-accuracy Inertial Measurement Unit (IMU) and nonlinearity of the filtering model. The system takes the low-accuracy IMU/GPS tight combination model as the framework

the system state equation and measurement equation are derived

the extended Kalman filter(EKF) algorithm is used to correct the error of INS with real-time feedback

and finally a real-vehicle test platform to verify the automatic navigation system is built under low-speed scenarios. The test results show that the tight combination system proposed in this paper reduces the longitude

latitude

northward velocity

and eastward velocity errors (standard deviation) by 24%

47%

50% and 54% respectively

which proves that the tight combination can effectively suppress the navigation error dispersion

compared with the loose combination system; and the EKF also effectively improves the navigation accuracy and enhances the filtering effect compared with the Kalman filter (KF) algorithm.