Sensor Bias Estimation in Distributed Multi-UUV Bearing-Only Target Tracking
Xu Jian, Zhuoran Sun · 2025
Bearing-only estimation is a challenging task in distributed multi-UUV (Unmanned Underwater Vehicle, UUV) target tracking. And the system bias of the passive sensors will exacerbate the accuracy of bearing-only estimation. Therefore, in order to improve the accuracy of the estimation, system bias estimation and registration is an essential process. In this paper, after constructing the error pseudo-measurement equation in the Cartesian coordinate system, in response to the problem of that the recursive least squares method is prone to data saturation, a real-time bias estimation algorithm based on weighted forgetting factor recursive least squares (WFFRLS) is carried out. And a real-time bias registration is realized by using extended Kalman (EKF) filter. The simulation comparison results show that the algorithm can estimate the system error more accurately than the classical recursive least squares method, and effectively enhanced the accuracy of bearing-only estimation. In addition, the bias estimation result is very close to the Cramer-Rao Lower Bound (CRLB), which shows that the proposed algorithm is efficient.