Attitude-aware method for hydraulic support groups in a virtual reality environment

Current monitoring methods for hydraulic supports are designed for single machines and are unsuitable for global monitoring. This paper presents an attitude-aware method for hydraulic support groups in a virtual reality environment, which predicts the next cycle attitude in real time by a Grey–Markov prediction model. The attitude datum includes the supporting height and the corresponding attitude angles, which are obtained by installing tilt sensors on the base, the front or rear connecting rod, or the roof beam. The operation state of the hydraulic supports was determined by comparing the predicted and actual data. Next, the hydraulic support attitude was related to the coal-mining height observed through multiple cycles. The three-dimensional virtual monitoring was solved in SQL SERVER, Kingview, Unity3d, and Matlab software. Finally, the proposed method was verified on 40 hydraulic supports of a working face through 17 cycles in an underground experiment. The predictive lateral and longitudinal accuracy reached 78.9 and 82.3%, respectively. By efficiently visualizing the operation state of the hydraulic support groups in three-dimensional mode, this method can predict the future attitude with high accuracy and reliability. Therefore, it provides a theoretical approach to safe and efficient operation of a fully mechanized coal-mining face.