DOO-Driven Sensor Selection: An Adaptive Framework for Underwater Multi-Sensor Navigation Systems
Ziyi Wang, Xue Du, Zhao Wang · 2025
The lack of standardized sensor performance metrics has made sensor selection algorithms to a persistent challenge in multi-sensor navigation system. To address this problem, this paper proposes an degree-of-observability (DOO)-driven multi-sensor selection framework. The DOO is utilized to quantify sensor measurements, addressing the lack of a quantitative index for sensor performance. Furthermore, by considering the adaptability between sensor types and environmental conditions, sensors with significant errors are excluded from the fusion process, resulting in an optimized sensor ensemble. Finally, the feasibility of the proposed algorithm is validated through semi-physical simulation.