In GNSS-denied environments, Micro-Electromechanical Systems–Inertial Navigation Systems (MEMS-INS) play a critical role in sustaining the autonomous operation of vehicles. However, the inherent accuracy constraints of MEMS inertial sensors often hinder their broader application. To address this limitation, this paper proposes a novel integrated navigation system based on a rotating MEMS architecture. Through a global observability analysis, the proposed method enables the effective separation of sensor biases, mitigates long-term error accumulation, and significantly enhances the estimation accuracy of both attitude and velocity. Simulation results indicate that a rotation scheme around the X-axis allows for the reliable estimation of sensor biases. Moreover, road test results reveal that when the vehicle experiences variations in angular velocity, the rotating configuration consistently outperforms its fixed counterpart, reducing heading, position, and velocity errors by 75.3%, 62.9%, and 76.0%, respectively.
Yang et al. (Thu,) studied this question.