具有丢包补偿的分布式一致性融合估计器

Consensus-Based Distributed Fusion Estimatior with Compensations of Packet Dropouts

摘要:

研究了具有随机丢包的网络化分布式一致性估计问题。丢包现象存在于各节点间局部状态估计值的传输过程中,引入一组服从 Bernoulli分布的随机变量来描述。当发生丢包时,以融合节点前一时刻融合估计值的一步预测值进行补偿。建立了以估计器增益为决策变量,以所有传感器有限时域下状态融合估计误差和为代价函数的优化问题。在给定一致性权重下,通过最小化代价函数的上界得到了一组次优的估计器增益,并给出了融合估计器渐进稳定的充分条件。最后,通过算例仿真验证了算法的有效性。

Abstract:

In this paper, the networked consensus-based distributed estimation problem with random packet dropouts wasstudied. The phenomenon of packet losses exists in the transmission of local state estimation values between sensor nodes,which wer described by a group of Bernoulli distributed random variables. The one-step prediction values of the fusion es.timate at the previous time of the fusion node were used as the compensation in the presence of packet losses. The optimi.zation problem was established by using the estimator gain as the decision variable and the state fusion estimation errorsand the cost function of all sensors in the finite time domain. Given a consistent weight, a set of suboptimal estimator gainswere obtained by minimizing the upper bound of the cost function, and a sufficient condition for the asymptotic stability ofthe fusion estimator was given. Finally, the effectiveness of the algorithm was verified by a numerical example.