A modified adaptive Kalman filtering method for maneuvering target tracking of unmanned surface vehicles. (15th December 2022)

• Record Type: Journal Article
• Authors: Fan, Yunsheng; Qiao, Shuanghu; Wang, Guofeng; Chen, Si; Zhang, Haoyan
• Is Part Of: Ocean engineering. Volume 266(2022) Part 3
• Publication Date: 2022-12-15
• Subjects: Adaptive Kalman filter -- Target tracking -- Position -- Velocity -- Radar -- Unmanned surface vehicle; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2022.112890
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: The filtering methods are crucial for an unmanned surface vehicle (USV) to realize target tracking. Due to the poor observation caused by the strong vibration during the navigation of the USV, the target tracking accuracy of the traditional filtering method has been significantly degraded. A modified strong tracking-based expended Sage-Husa adaptive robust Kalman filter (MST-ESHARKF) algorithm is proposed to overcome this problem in this paper. In the proposed algorithm, a modified fading factor for the strong tracking Kalman filter is introduced to eliminate disturbance-induced filter divergence. In addition, the adaptive factor of robust Kalman filtering is designed to balance the predicted and observed states dedicated to improving the robustness of the algorithm. Finally, the biased and unbiased estimators for measurement and process noise covariances are merged, and the measurement noise covariance matrix's interval is constrained, resulting in a simultaneous evaluation of measurement and process noise covariance matrices with improved dependability of the proposed algorithm. The simulation and experiment results show that the proposed MST-ESHARKF outperforms the existing filters in target tracking. Highlights: To improve the target state's estimation accuracy under strong vibration, a modified strong tracking-based expended Sage-Husa adaptive robust Kalman filter is proposed. The fading factor of the strong tracking Kalman filter is modified to correct the one-step prediction error covariance matrix to prevent filter divergence. A robust Kalman filter with a modified adaptive factor is developed to eliminate the effects of outliers. The measurement noise range is limited to a minimum and maximum threshold, after which the unbiased and biased noise estimators are integrated to improve the algorithm's reliability. Strong tracking and robust Kalman filtering are combined with Sage-Husa adaptive Kalman filtering to improve the algorithm's stability and robustness.