Research on Matching Accuracy of Underwater Terrain Matching Algorithm Based on Unscented Kalman Filter
Lu Xiong, Shen Jian, Bi Xiaowen · 2019
UT transform is the core step of unscented kalman filter (UKF) algorithm. The Sigma sampling type of UT transform has great influence on the computational complexity, state estimation accuracy and computational efficiency of UKF-based terrain matching algorithm. In order to obtain the high precision terrain matching method which is suitable for engineering application, the Comparative analysis of three kinds of UKF-based terrain matching algorithms was carried out in this paper. Firstly, the UKF-based underwater terrain matching model was given. Secondly, the principle and the performance characteristics of scaled sampling method, Cubature sampling method, Gauss-Hermite sampling methods were analyzed. The filtering precision factors of symmetric unscented Kalman filter(SUKF), Cubature unscented Kalman filter(CKF), Gauss-Hermite Kalman Filter(GHKF) was studied from the algorithm principle, Thirdly, The computer simulation was performed using univariate nonstationary growth model(UNGM) to test the previous analysis conclusions. Finally underwater terrain matching simulation model was established, and the comparative analysis of matching acuraccy was carried out based on SITAN, SUKF, CKF and GHKF, the matching effect of underwater terrain matching algorithm is obtained. The simulation results show that the UKF algorithm based on symmetric sampling strategy has better filtering precision and less computational complexity which is suitable for the filtering core of underwater terrain matching algorithm.