Abstract: In terms of the problem that the straightness detection of scraper conveyor is difficult and the measurement accuracy is not high at a fully mechanized coal face,a new method of scraper conveyor shape detection,which is based on the size of central shute and the dead reckoning,is proposed using SINS. Firstly,the INS attitude is calculated by Rodrigues parameter algorithm. Then,the vibration signal can be obtained using the IMU when shearer passes the con- nection portion of scraper conveyor and the section number detection model of scraper conveyor can be established with Butterworth filter. At last,the shape detection model of scraper conveyor is built with the consideration of connection gap. An experimental platform is built both in horizontal and vertical plane with the models of shearer and scraper con- veyor. Furthermore,the results of experiment show that the shape tracking error of scraper conveyor is less than 14 mm. The shape detection method proposed provides a new way for the shape detect of scraper conveyor,which provides a theoretical basis for the implementation of automatic alignment at the fully mechanized coal face.