一种无锚节点的协同定位算法

A cooperative localization algorithm for anchor-free nodes

为了解决无锚节点协同定位过程中累积误差的传播问题,提出了一种基于mesh网络改进的map-growing算法。该算法首先通过推导参考节点与待测节点的定位误差估计值,提出了基于几何的虚拟锚节点共线度表达式;再根据多条件约束确定了虚拟参考节点建立相对参考坐标系,并实现了最优节点集定位;最后引入节点不确定度实现节点升级策略,控制迭代定位过程中累积误差的扩散。在不同节点分布场景下的仿真结果表明,所提算法有效的缓解了协同定位过程中的累积误差,取得了较好的定位性能。

In order to solve the problem of continuous error accumulation caused by propagation in the process of anchor-free cooperative localization,an improved map-growing algorithm based on mesh network is proposed.Firstly,the geometric collinearity formula of virtual anchor node is presented by deducing the estimation value of positioning error between reference nodes and node to be tested.Then,the virtual reference nodes are selected according to the multi-condition constraints to establish the relative reference coordinate system,and the optimal node set is used to complete the cooperative localization.Finally,the node uncertainty is introduced to accomplish the node upgrading strategy,which can curb the spread of cumulative error in the iterative positioning process.Simulation results in various node distribution scenarios show that the proposed algorithm effectively alleviates the accumulated errors in the collaborative positioning process and achieves good positioning performance.