To address the challenge of maintaining accurate and reliable height control of shearers’ rocker arms under underground mining conditions, this study proposes an improved method for detecting inclination angles using a MEMS accelerometer. Traditional detection methods, such as the cylinder stroke displacement and coded potentiometer rotation ranging techniques, are prone to decreased accuracy and reliability due to long-term wear on the rocker hinge shafts and difficult maintenance. In this work, we introduce filtering strategies designed to mitigate high-frequency and high-amplitude vibration noise encountered in harsh vibration environments, thereby enhancing measurement accuracy. Specifically, the critical damping method and combined integration approach are employed to process the raw triaxial data from the accelerometer, effectively isolating and extracting useful gravitational acceleration data to determine the angle. A simulation experiment platform was constructed to replicate the vibration conditions experienced by the rocker arm. Through this platform, dynamic inclination angle identification within a vibrating environment is achieved, significantly improving angle measurement accuracy. The experimental results indicate that in a 5g vibration environment, both filter designs exhibit faster response speeds and can rapidly track changes in the input signal. After applying the combined integral filter, the angle error is less than 0.3°, and after the critical damping filter application, the angle error is reduced to less than 0.1°. This level of precision satisfies the actual demand for controlling the mining height of the rocker arm. The proposed method provides a feasible solution for detecting the inclination angle of the shearer’s rocker arm, offering enhanced accuracy and reliability without being affected by mechanical wear or maintenance challenges, thus contributing to safer and more efficient mining operations.