基于单水声信标定位的海底飞行节点群体定位研究

Ocean bottom flying node (OBFN) is a scalable, unique alternative to currently implemented solutions for ocean bottom seismic observation of petroleum resources. The OBFN can cruise to the designated location down to the depth of 3000 meters, and remain static at the seafloor to acquire long term seismic data for up to 30 days. Accurate positioning is essential to ensure the accuracy of the gathered data. The localization of OBFNs is a challenging problem since the amounts of OBFNs need to be deployed can be up to 10 thousands. In this project, the localization of OBFNs will be achieved by extension of traditional single beacon underwater tracking system for single underwater agent. Necessary and sufficient conditions on the observability of a novel single beacon tracking model with unknown effective sound velocity will be derived. Subsequently, non-linear Kalman filter based on Interacting Multiple Model concept, which combines the extended Kalman filter, unscented Kalman filter and Particle filter, will be developed to solve the problem of large linearization error while implementing commonly used extended Kalman filter. In addition, an improved Sage-Husa adaptive filter will be designed to account for the unknown time varying probability distribution of process and measurement noise. Furthermore, to improve positioning accuracy of OBFNs, an unmanned surface vehicle (USV) will be designed to carry the acoustic beacon and periodically broadcast its position to the OBFN network. The path planning algorithm for the USV aiming at cooperative USV/OBFN navigation will be developed. This project will not only provide a novel solution method for localization of OBFNs, but also promote the development of ocean bottom seismic research.

海底飞行节点（OBFN）因具有信号采集质量好、布放回收效率高、地形适应性强等优势，是海底石油勘探技术的研究热点。针对OBFN大规模组网布放面临的群体定位问题，本项目拟将针对单一目标的单水声信标定位方法扩展到水下节点群体定位中，通过将非线性定位模型线性化，研究基于未知声速的单水声信标定位模型局部能观测性及全局能观测性；基于交互式多模型思想，融合粒子滤波、无迹卡尔曼滤波等非线性滤波算法，研究针对OBFN群体定位的非线性滤波算法；通过设计动态调节反馈窗口宽度的模糊控制器，研究改进的Sage-Husa自适应滤波算法；以最小化群体定位误差为目标函数，研究保证群体定位能观测性的信标载体动态路径规划算法。遵循从模型到算法、从单一到群体的研究思路，解决单水声信标定位非线性状态估计、时变系统噪声参数估计、USV/OBFN协同定位等关键科学问题，为实现OBFN大规模组网布放提供位置保障。

海底飞行节点（OBFN）因具有信号采集质量好、布放回收效率高、地形适应性强等优势，是海底石油勘探技术的研究热点。针对OBFN大规模组网布放面临的群体定位问题，本项目将针对单一目标的单水声信标定位方法扩展到水下节点群体定位中，通过将非线性定位模型线性化，研究了基于未知声速的单水声信标定位模型局部能观测性及全局能观测性；基于交互式多模型思想，融合粒子滤波、无迹卡尔曼滤波等非线性滤波算法，提出了针对OBFN群体定位的非线性滤波算法；通过设计动态调节反馈窗口宽度的模糊控制器，提出了改进的Sage-Husa自适应滤波算法；融合长基线与单信标定位方法，提出了基于VB近似的水下鲁棒自适应群体定位方法。所提出的水下单信标定位方法成功应用于团队自主研发的海底飞行节点AUV等多型水下自主潜器中，基于外场湖试实测结果表明，所提方法将单水声信标定位系统的定位误差由10m量级降低到m级，定位精度显著提高。项目研究工作在2019年1月至2021年12月按计划顺利开展，研究成果抵达预期，完成了既定的研究目标。在本项目支持下，项目组以第一作者或通讯作者共发表论文5篇，其中SCI检索论文4篇，EI检索论文1篇；授权国家发明专利5项，受理国家发明专利12项；出版英文专著（Book Series) 1部。

内容获取失败，请点击重试