1. 基于最大互相关熵UKF的传感网目标状态和系统偏差联合估计算法.
- Author
-
赵季红, 谢志勇, 曲桦, 王明欣, and 刘熙
- Subjects
- *
SENSOR networks , *MEASUREMENT errors , *TRACKING algorithms , *NONLINEAR equations , *COVARIANCE matrices , *EQUATIONS of state , *VECTOR error-correction models , *KALMAN filtering - Abstract
This paper proposed a method that combined target state and system error based on the maximum correntropy criterion unscented Kalman filter(MCUKF), namely ASMCUKF, which could solve the problems that the observation noise of sensor was heavy-tailed or had some sudden change in space-air-ground integrated sensor network. Firstly, MCUKF used unscented transformation(UT) to obtain the predicted state estimation and covariance matrix, and then used a non-linear regression model based on MCC to reconstruct the observed information and strengthen the robustness of UKF in heavy-tailed noise. The ASMCUKF algorithm established the state equation and the nonlinear observation equation with systematic errors through the expansion of target state vector, and made the error registration according to the estimated system errors to improve the influence of the system errors on the target state estimation. The simulation results show that the ASMCUKF has a better performance for the state estimation of the communication target than the traditional method in the environment of heavy-tailed non-Gaussian noise. [ABSTRACT FROM AUTHOR]
- Published
- 2019
- Full Text
- View/download PDF