基于多个自主水下航行器的分布式协同流场估计

Distributed cooperative flow field estimation using multiple AUVs

中文关键词  流场估计  树型结构  自主式水下航行器  分布式算法  非线性Kaczmarz

英文关键词  flow field estimation  tree structure  autonomous underwater vehicle  distributed algorithm  nonlinear Kaczmarz

中文摘要

      本文考虑利用多个自主式水下航行器(AUV)实现流场估计, 提出了一种基于树型网络的分布式方法来估计 水下流场. 在本文中, 借助绝对运动积分误差和相对运动积分误差, 流场估计问题被描述为求解一个以未知流场为 变元的非线性方程组. 继而本文在多AUV系统内建立一个低通讯成本的树型网络, 并在该网络上运行一种分布式 算法以求解与流场估计相关的非线性方程组. 在该算法中, 每个AUV将当前的流场估计值连续地投影到自身拥有 的约束方程的解集中, 并通过扩散和池化两个步骤在树型网络间传递流场估计值. 本文证明了上述算法的收敛性, 并通过仿真实验验证了所述分布式协同流场估计方法的有效性.

英文摘要

      This paper considers the use of multiple autonomous underwater vehicles (AUVs) to estimate the flow field, and proposes a distributed method based on a tree-structure network to estimate the underwater flow field. In this paper, the flow field estimation problem is described as solving a nonlinear equation system with the unknown flow field as argument using absolute motion-integration error and relative motion-integration error. Then this paper establishes a tree-structure network with low communication cost in the multi-AUV system to run a distributed algorithm to solve the nonlinear equation system related to the flow field estimation. In this algorithm, each AUV continuously projects current flow field estimation value to the solution set of its own constraint equations. And through the two steps of dispersion and pooling, the flow field estimation value is transferred within the tree-structure network. This paper proves the convergence of the proposed algorithm, and the effectiveness of the distributed cooperative flow field estimation method is verified through simulation experiments.