基于最小一致性的自治式水下机器人路径规划

Path Planning of Autonomous Underwater Vehicle via Minimum Consensus

鉴于现有路径规划方法难以兼顾路径质量与计算效率,采用改进的最小一致性算法,用于解决海洋环境下的自治式水下机器人(AUV)路径规划问题。首先,在传统最小一致性算法的基础上,引入相邻节点间的连接权重以计算各节点的控制输入,迭代更新节点状态值并最终到达稳定状态。然后,根据离散化的栅格空间与多层邻居栅格策略构建节点图,并基于能量消耗与平滑度指标确定各节点状态值及连接权重,实现AUV全局路径规划。此外,由于AUV沿全局路径航行时可能遭遇突发威胁,因此引入偏离度等指标重新定义节点图,重规划AUV局部路径。最后,仿真结果表明了本文方法的有效性。

As the existing methods have a trade-off between path quality and calculation efficiency,this paper improves the minimum consensus method to solve the path planning problem of autonomous underwater vehicle(AUV)in ocean environment.First,the edge weight is introduced into the standard minimum consensus method to calculate the control inputs and update the states of nodes,and the system will finally converge to stability.Then,in order to achieve the global path planning,the graph is constructed via the discretized grids and multi-layer neighbors,where the node states and edge weights are determined on the basis of the index of energy consumption and smoothness.In the case that AUV encounters pop-up threats when travelling along the global path,the graph is redefined to generate the local path by introducing the path deviation.Finally the simulation results verify the feasibility of our method.