空中无人集群时滞置信传播协同导航算法

Collaborative Navigation Algorithm for Delay Belief Propagation of Unmanned Aerial Cluster

空中无人集群协同导航系统依赖于载体间的信息传输,信息传输时间导致相对导航观测出现时滞。针对传统置信传播算法消息传递时未考虑观测时滞的问题,提出了一种空中无人集群时滞置信传播协同导航算法。首先根据置信传播中相对测距信息轮播出现的时滞量建立时滞功能节点误差模型,再通过卫星伪距一阶泰勒展开项与导航卫星运动速度的数量级变化推算伪距值,并利用时滞置信传播算法完成所有无人飞行器的后验概率估计。仿真结果表明,无人飞行器定位精度在60 m/s的情况下最高提升了32.70%,在100 m/s的情况下最高提升了48.01%,满足了中高速飞行状态下的定位需求,有效地改善了传统置信传播算法的协同导航性能。

The unmanned aerial cluster collaborative navigation system relies on information transmission between carriers.Information transmission time causes a time-delay in relative navigation observations.For traditional belief propagation(BP)algorithms,message delivery does not consider the problem of observation time-delay.In this paper,a time-delay collaborative navigation algorithm for unmanned aerial cluster based on belief propagation is proposed.Firstly,the timedelay function node error model is established according to the time-delay amount of relative ranging information carousel in belief propagation.Then,the pseudorange value is estimated from the order of magnitude change between the first-order Taylor expansion of the satellite pseudorange and the motion speed of the navigation satellite.Finally,the delay belief propagation algorithm(DBPA)is used to estimate the posterior probability of all unmanned aerial vehicles.The simulation results show that the positioning accuracy of the unmanned aerial vehicle is increased by up to 32.70%at a speed of 60 m/s and by up to 48.01%at a speed of 100 m/s.It meets the positioning needs in medium and high speed flight conditions.The collaborative navigation performance of the traditional belief propagation algorithm is effectively improved.