Underwater Target Tracking Based on Strong Tracking Sparse Grid Quadrature Filter

摘要

Underwater moving object detection/tracking iscritical in various applications such as exploration of naturalundersea resources, acquiring of accurate scientific data tomaintain regular surveillance of missions, navigation andtactical surveillance. In currently, underwater moving target isusually tracked using the traditional non-linear estimators suchas Extended Kalman Filter (EKF) and unscented Kalman Filter(UKF). However, if an underwater target moves with delicatemaneuver, the accuracy of the filter may decline, even diverge.In this paper, a (STSGQF) is proposed to deal with the problem.The STSGQF is obtained by introducing the Strong TrackingFilter (STF) to the Sparse Grid Quadrature Filter (SGQF).Compared with the Gauss-Hermite Quadrature Filter (GHQF),the sparse grid method is available to reduce the SGQF'scomputational cost significantly, with slight sacrifice ofaccuracy its accuracy declines slightly. Meanwhile, theSTSGQF has stronger robustness than SGQF against the statechange. The effectiveness of STSGQF is demonstrated by thesimulation results more robust, better robustness.