针对自主水下航行器（AUV）在统计特性时变且未知的高斯噪声条件下作业时的定位精度不足问题，提出一种基于自适应阈值与改进Sage-HusaUKF（Adaptive Threshold and Improved Sage-HusaUKF，AT-ISHUKF）的AUV 协同定位算法。首先，对量测噪声协方差矩阵更新进行改进，以降低误差协方差矩阵失去正定性的风险；然后，引入自适应阈值判断噪声特性变化是否明显，若变化明显，采用较小的自适应因子来更快地适应这种变化，若变化不明显，则采用较大的自适应因子来保持估计的稳定性。仿真结果表明，在噪声统计特性时变且未知的基础上，统计特性突变为原来的5倍，平均定位误差比Sage-Husa-UKF算法减少了48.9%，均方根误差减少了61.7%。提出的AT-ISHUKF算法在噪声统计特性时变且未知的情况下有效地提高了跟随AUV的定位精度和滤波的鲁棒性。
Aiming at the problem that the positioning accuracy of autonomous underwater vehicle (AUV) is insufficient when it works under the condition of unknown Gaussian noise with time-varying statistical characteristics, an AUV cooperative location algorithm based on adaptive threshold and improved Sage-HusaUKF (AT-ISHUKF) is proposed. Firstly, the measurement noise covariance matrix update is improved, so as to reduce the risk that the error covariance matrix loses its positive definiteness. Then, adaptive threshold is introduced to judge whether the noise characteristic changes obviously, if the change is obvious, a smaller adaptive factor is adopted to adapt to this change more quickly, if the change is not obvious, a larger adaptive factor is adopted to maintain the stability of the estimation. The simulation results show that, on the basis that the statistical characteristics of noise are time-varying and unknown, the statistical characteristics suddenly change to five times of the original, and the average positioning error is reduced by 48.9% and the root mean square error is reduced by 61.7% compared with Sage-Husa-UKF algorithm. The proposed AT-ISHUKF algorithm effectively improves the positioning accuracy and filtering robustness of following AUV when the statistical characteristics of noise are time-varying and unknown.
2025,47(19): 107-114 收稿日期：2024-12-27
DOI：10.3404/j.issn.1672-7649.2025.19.017
分类号：U674.94
基金项目：国家自然科学基金资助项目（62361018）
作者简介：陈辉(1976-),男,博士,副教授,研究方向为光通信、水下导航定位技术
参考文献：
[1] 闫敬, 陈天明, 关新平, 等. 自主水下航行器协同控制研究现状与发展趋势[J]. 水下无人系统学报, 2023, 31(1): 108-120.
[2] 刘峰, 王宇雄, 陈惠芳, 等. 集成逆超短基线的自主式水下航行器集群协同定位方法[J]. 声学学报, 2023, 48(4): 687-698.
[3] 李倩, 聂简, 黄鸿殿, 等. 基于大脑海马认知机理的主从式AUV协同定位方法[J]. 中国惯性技术学报, 2024, 32(1): 27-33.
[4] 何桂萍, 王正伟, 陈洲. 基于自适应残差加权的AUV集群分布式协同定位算法[J]. 舰船科学技术, 2022, 44(18): 101-105+189.
HE G P, WANG Z W, CHEN Z. Distributed cooperative localization algorithm of AUV cluster based on adaptive residual weighting[J]. Ship science and technology, 2022, 44(18): 101-105+189.
[5] 陈世杰. 主从式多AUV系统协同定位算法研究[D]. 南京: 东南大学, 2023.
[6] JIANG L, GAO W G, LI Y F, et al. Cooperative localization for master-salve multi-AUVs based on range measurements [J]. Physical Communication, 2023, 61.
[7] JIANG L Y, LI Y C, YU W B, et al. Cooperative localization for asynchronous AUVs using time difference of communication in underwater anchor-free environments[J]. IEEE Transactions on Cybernetics, 2024, 54(11): 6531-6544.
[8] ZHANG F B, WU X Q, MA P. Consistent extended Kalman filter-based cooperative localization of multiple autonomous underwater vehicles[J]. Sensors, 2022, 22(12): 4563-4563.
[9] 肖鹏飞, 许至尊, 白虎林, 等. 自适应无迹卡尔曼滤波算法在水下组合导航系统中的应用[J]. 广东海洋大学学报, 2024, 44(4): 121-128.
[10] 周萌萌, 张冰, 赵强, 等. 基于自适应渐消 Sage-Husa 扩展卡尔曼滤波的协同定位算法[J]. 中国舰船研究, 2022, 17(4): 92-97.
[11] 张淋, 谭良成. 水下航行器自主导航定位技术前沿进展[J]. 兵器装备工程学报, 2024, 45(3): 161-171.
