摘要
无人机自主决策是无人机自主控制技术的关键技术之一。为提高静态贝叶斯、动态贝叶斯等传统方法环境适应能力,针对参数随时间平稳变化的情形,提出了一种基于参数时变离散动态贝叶斯网络的粒子滤波参数估计方法。形式化描述了结构不变条件下的参数时变离散动态贝叶斯网络的一般模型;给出了PTVDDBN参数估计与推理决策框架;提出了基于粒子滤波的PTVDDBN参数估计方法;以战场环境中无人机协同行为决策为背景,开展了时变参数估计及PTVDDBN模型推理决策试验。通过对比静态贝叶斯、动态贝叶斯等方法,可知PTVDDBN模型算法能够更加准确地估计时变参数,推理决策准确性更高。
UAV autonomous decision-making is one of the key technologies of UAV autonomous control technology.In order to improve the environmental adaptability of traditional methods such as Bayesian network and dynamic Bayesian network,aiming at the situation where the parameters change smoothly with time,a parameter estimation method of particle filter based on the parameter time-varying discrete dynamic Bayesian network(PTVDDBN)is proposed.Under the condition of invariable structure,the general model of the discrete dynamic Bayesian network with time-varying parameters is described formally.A novel method of parameter estimation and inference decision framework of PTVDDBN is put forward.A parameter estimation method of PTVDDBN based on particle filter is proposed.With UAV cooperative decision-making in battlefield as the application background,time-varying parameter estimation and PTVDDBN model inference and decision-making experiments are carried out.By comparing static Bayesian network and dynamic Bayesian network models,it is concluded that the PTVDDBN model algorithm can estimate the time-varying parameters more accurately,and the decision-making inference is more reliable.
作者
吴英捷
曹欣芹
李杰
Wu Yingjie;Cao Xinqin;Li Jie(College of intelligence,National University of Defense Technology,Changsha 410073,China;Unit 78092 of PLA,Chengdu 610000,China)
出处
《战术导弹技术》
北大核心
2020年第6期53-59,139,共8页
Tactical Missile Technology
关键词
参数时变离散DBN
贝叶斯网络推理
参数估计
粒子滤波
无人机协同行为决策
parameter time-varying discrete dynamic Bayesian network
Bayesian network inference
parameter estimation
particle filtering
UAV cooperative decision-making
