摘要
为了提高移动传感器网络时延系统控制能力,提出基于强化学习的移动传感器网络时延系统控制模型,采用高阶近似微分方程构建移动传感器网络时延系统的控制目标函数,结合最大似然估计方法进行移动传感器网络的时延参数估计,采用强化学习方法进行移动传感器网络的收敛性控制和自适应调度,建立传感器网络时延系统控制的多维测度信息配准模型,在强化跟踪学习寻优模式下实现移动传感器网络时延系统的自适应控制。仿真结果表明,采用该方法进行移动传感器网络时延系统控制的自适应性较好,时延参数估计准确度较高,控制过程的鲁棒性较强。
In order to improve the control ability of time-delay systems in mobile sensor networks,a control model of time-delay systems in mobile sensor networks based on reinforcement learning is proposed.The control objective function of timedelay systems in mobile sensor networks is constructed by using high-order approximate differential equations.The maximum likelihood estimation method is used to estimate the time delay parameters of mobile sensor networks,and the reinforcement learning method is used to control the convergence and adaptive scheduling of mobile sensor networks.A multi-dimensional measure information registration model for time-delay system control in sensor networks is established,and the adaptive control of time-delay systems in mobile sensor networks is realized in the mode of intensive tracking and learning optimization.The simulation results show that the proposed method has better self-adaptability,higher accuracy of time-delay parameter estimation and stronger robustness of the control process for time-delay systems in mobile sensor networks.
作者
杨雯雯
徐小玲
YANG Wenwen;XU Xiaoling(Public Teaching Department of Yan'an Vocational and Technical College,Yan'an Shanxi 716000,China;Xi'an Innovation College of Yan'an University,Xi'an 71000,China)
出处
《自动化与仪器仪表》
2020年第2期62-65,共4页
Automation & Instrumentation
基金
陕西省教育科学规划课题:基于大数据的教学质量评价与分析研究(No.SGH18H466)
关键词
强化学习
移动传感器网络
时延
系统控制
reinforcement learning
mobile sensor networks
delay
system control
