通信随机时滞条件下基于分布式模型预测的AUV编队控制  

AUV formation control with communication stochastic delay based on distributed model prediction

作　　者：徐博[1,2] 王朝阳 王潇雨[1] 沈浩 XU Bo;WANG Zhao-yang;WANG Xiao-yu;SHEN Hao(College of Intelligent Systems Science and Engineering,Harbin Engineering University,Harbin 150006,China;Science and Technology on Near-Surface Detection Laboratory,Wuxi 214000,China)

机构地区：[1]哈尔滨工程大学智能科学与工程学院,哈尔滨150006 [2]近地面探测技术重点实验室,江苏无锡214000

出　　处：《控制与决策》2023年第5期1363-1372,共10页Control and Decision

基　　金：国防科技创新特区项目(19-163-16-ZD-020-002-01);国防科技卓越青年科学基金项目(2020-JCJQZQ-071);国防基础科研计划项目(JCKY2019604D003);近地面探测技术重点实验室基金项目(6142414211405)。

摘　　要：考虑水声通信随机时滞条件下AUV编队协同控制问题,提出一种基于分布式模型预测的AUV编队控制方法.首先,通过所设计的随机时滞通信同步策略,将异步状态信息转换为同步状态信息;然后,结合虚拟轨迹、状态预测、控制约束以及编队内AUV状态信息描述协同编队代价函数,将其引入局部滚动时域优化,实现编队控制目标,并利用李雅普诺夫理论验证编队控制器的稳定性;最后,将所提出方法与现有编队控制方法进行对比仿真,仿真结果验证了其有效性.Considering the problem of AUV formation cooperative control in underwater acoustic communication with random time delay,this paper proposes a distributed model prediction based AUV formation control method.Firstly,the asynchronous state information is transformed into synchronous state information by the designed random time-delay communication synchronization strategy,and then the cooperative formation cost function is described by combining virtual trajectory,state prediction,control constraints and AUV state information.The formation control function is introduced into local rolling time domain optimization to achieve formation control objectives.The Lyapunov theory is used to prove the stability of formation controller.Finally,the proposed method is compared with the existing formation control methods,and the results verify its effectiveness.

关 键 词：AUV 编队协同控制 通信随机时滞 虚拟轨迹 分布式预测控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]