基于节点通信度的信息加权一致性滤波

协同目标跟踪是无人机集群等多传感器网络的典型应用。在分布式传感器网络目标跟踪过程中,目标状态估计的一致性直接影响到跟踪有效性。针对目标跟踪过程中网络节点之间一致性迭代次数受限的问题,提出了一种基于节点通信度的信息加权一致性滤波算法,设计了用节点通信度来充分衡量传感器节点在网络中的通信拓扑状况,并构建了非对称一致性权值的选取机制,可在复杂拓扑结构网络中实现快速一致性跟踪。典型目标跟踪场景仿真验证表明,所提算法相比经典的信息加权一致性滤波算法,目标跟踪的不一致程度降低了20%以上,有效提升了分布式跟踪的一致性速度。

Information weighted consensus filter algorithm based on node communication degree

Collaborative target tracking is one of the most typical applications of multi-sensor network including unmanned aerial vehicle swarm. In the process of target tracking in distributed sensor network, the consensus of target state estimation directly affect the tracking effectiveness. Aiming at the problem of limited consensus iteration number among network nodes in the process of target tracking, an information weighted consensus filtering algorithm based on node communication degree is proposed. The node communication degree is designed to fully measure the communication topology of sensor nodes in the network. And then, a selection mechanism of asymmetric consensus weights is constructed to achieve fast consensus tracking in the complex topology network structure. It is verified by the simulation of typical target tracking scene, compared with the classical information weighted consensus filtering algorithm, the proposed algorithm reduces the degree of disagreements of target tracking by more than 20%, and effectively improves the consensus speed of distributed tracking.