Velocity-Based Optimization-Based Alignment (VBOBA) of Low-End MEMS IMU/GNSS for Low Dynamic Applications
Quan Zhang, Shanshan Li, Zhengpeng Xu, Xiaoji Niu
Abstract
To solve the low accuracy and poor reliability of the alignment of low-end micro-electro-mechanical systems (MEMS) inertial measurement units (IMU) in low dynamic applications such as precision agriculture, this paper proposes a velocity-based optimization-based alignment (VBOBA) method using the assistance of GNSS velocity. Essentially, the proposed VBOBA method is a kind of attitude determination algorithm based on multiple velocity vector observations. Furthermore, the global perspective of observability analysis is utilized to investigate the influence of vehicle maneuvers on the estimation of the heading angle, which reveals that the proposed VBOBA method only requires velocity vectors, while the existing acceleration-based optimization-based alignment (ABOBA) method requires acceleration changes to estimate the heading angle. The results of simulation and tractor tests show that the alignment error of the heading angle provided by the proposed VBOBA method can be reduced to 4° in 60s under low dynamic conditions with about 1 m/s for the industrial grade MEMS IMU (e.g., ADIS16460). The proposed VBOBA method can serve as an effective in-motion alignment for low-end and low dynamic applications.