At present, most of the underground mining equipment positioning methods use inertial navigation plus odometer, visual measurement, ranging radar and other auxiliary measurement methods to reduce the position error drift accumulated by inertial navigation over time. However, these methods requires high reliability and measurement accuracy of the auxiliary measurement methods, and the centralized Kalman filter is used to filter and solve the data. There are problems such as a large amount of calculation and poor fault tolerance. In order to solve the above problems, a continuous shearer positioning method consisting of strapdown inertial navigation, ranging radar and odometer is proposed, using the ranging radar and odometer to measure the displacement of the continuous shearer respectively. The dead reckoning method is used to calculate the two-dimensional position information of the machine body, and the two-dimensional position information is applied to correct the inertial navigation position solution information so as to reduce the position drift error accumulated by inertial navigation over time. The federated Kalman filter is used to filter, solve and fuse the position information derived from each sensor, and reduce the calculation amount by reducing the dimension of the equation. Therefore, this method ensures the measurement accuracy while having a certain degree of fault tolerance. The simulation results show that the positioning accuracy of this positioning method can reach ±2 cm, and the orientation accuracy can reach 15′. The federated Kalman filter algorithm has small calculation amount and certain fault tolerance, which can lay the foundation for accurate measurement of the position of the continuous shearer in the roadway constrained environment, automatic roadway forming control and remote control.