VINS-Integrity: An Integrity Monitoring Scheme for Visual/IMU Integration Based on Factor Graph Optimization
Yunong Tian, Xiaoyang Yu, Qiufang Lin · 2024
Visual/Inertial Measurement Unit (IMU) integrated navigation has garnered significant attention for its superior performance in complex environments. A great deal of research has focused on how to improve the accuracy of Visual/IMU integrated navigation in complex environments, while insufficient attention has been paid to its navigation integrity in the presence of faults. Integrity is an important safety metric for evaluating navigation performance. Compared with the filtering methods, the Visual/IMU integrated navigation system based on factor graph optimization (FGO) has the advantage of higher accuracy but lacks an effective integrity monitoring scheme. In this paper, we propose an integrity monitoring scheme for FGO-based Visual/IMU integration, including Fault Detection and Exclusion (FDE) and Protection Level (PL) capability. Simulation and field experiments were conducted to validate the effectiveness of the proposed scheme. Experimental results indicate that the devised FDE method effectively excludes faulty measurements, and the proposed PL overbounds the position error by 100%in all three fault modes.