Abstract: In view of the environmental constraints such as long working cycle and large vibration amplitude during the cutting of the roadheader in the fully mechanized excavation face, it is difficult for a single inertial navigation system to realize autonomous, real-time and accurate perception of the attitude and position of the roadheader in the fully mechanized excavation process due to the influence of navigation error accumulation with time. Taking the inertial navigation system as the main system, a combined positioning and orientation method based on inertial navigation system and laser sensing system is proposed. The inertial navigation system is used to obtain the real-time attitude information of the roadheader, which is transmitted to the laser sensing system and combined with its detection light point feature information. The spatial coordinate transformation is used to solve the position information, and the inertial navigation system is transmitted back to the inertial navigation system for pose auxiliary calibration. By integrating the advantages and disadvantages of each detection subsystem, the accurate detection of the pose of the roadheader is realized, which effectively overcomes the problems of single inertial navigation system detection value drift and poor system reliability. Finally, through the ground simulation test and the underground industrial test, the effectiveness and detection accuracy of the detection system are verified under different working conditions. The results show that the average error of the lateral offset detection of the integrated positioning system is less than 10 mm, and the average error of the longitudinal detection is less than 20 mm, which can meet the accuracy requirements of the lower position detection under different working conditions. The introduction of advanced information fusion technology into coal mining technology provides theoretical guidance and practical support for the further development of unmanned and intelligent coal roadway excavation.