约束条件下基于满意模型预测控制的动基座水下无人航行器对接控制  

Moving base unmanned underwater vehicles docking control based on satisfactory model predictive control with a constrained condition

作　　者：张伟[1] 伍文华 滕延斌 宫鹏 唐业竞 ZHANG Wei;WU Wenhua;TENG Yanbin;GONG Peng;TANG Yejing(Science and Technology on Underwater Vehicle Laboratory, Harbin Engineering University, Harbin 150001, China;State Key Laboratory of Hydrodynamic National Defense Science and Technology, China Ship Scientific Research Center, Wuxi 2014082, China)

机构地区：[1]哈尔滨工程大学水下机器人技术重点实验室,黑龙江哈尔滨150001 [2]中国船舶科学研究中心水动力学国防重点实验室,江苏无锡214082

出　　处：《哈尔滨工程大学学报》2022年第1期108-118,共11页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(E1102/52071108).

摘　　要：针对水下无人航行器与运动母船对接过程中存在多种复杂约束,单一优化目标无法满足多目标同时最优的问题,本文通过将满意控制思想引入到模型预测控制中,设计了一种约束条件下的多目标多自由度的满意模型预测控制对接控制器。首先,对动基座水下无人航行器对接过程中的多种复杂条件进行详细分析。然后,将水下无人航行器模型按照满意思想进行分布式处理,得到北东位置、深度、纵倾和航向模型。最后,将多种复杂约束分别引入到分布式的水下无人航行器预测模型和目标函数中,进行优化求解。仿真结果表明:该方法在处理多种复杂约束问题上的快速性和稳定性。To address the problems that multiple complex constraints exist in the docking process of unmanned underwater vehicles(UUVs)to a moving mother ship and that a single optimization target cannot satisfy a simultaneous optimization of multiple degrees of freedom in this process,we introduce the idea of satisfactory control into model predictive control(MPC)and design a satisfactory multi-objective MPC docking controller with multiple degrees of freedom.First,a detailed analysis of the various complex conditions in the moving base UUV docking process is made.Then,the UUV model is processed in a distributed manner according to the satisfactory idea,and the northeast position,depth,pitch,and yaw model are obtained.Finally,a variety of complex constraints are separately introduced into the distributed UUV prediction model and objective function to optimize the solution.The simulation results show the rapidity and stability of this method in dealing with multiple complex constraints.

关 键 词：动基座水下无人航行器 对接 满意模型预测控制 预测模型 目标函数 滚动时域 回收母船 

分 类 号：TP23[自动化与计算机技术—检测技术与自动化装置]