系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (1): 250-261.doi: 10.12305/j.issn.1001-506X.2022.01.31

• 制导、导航与控制 • 上一篇    下一篇

基于自校正MPC的舰载机着舰控制技术

韩维1, 崔凯凯1, 刘洁2,*, 王昕炜3, 张勇1   

  1. 1. 海军航空大学航空基础学院, 山东 烟台 264001
    2. 军事科学院战争研究院, 北京 100850
    3. 大连理工大学工程力学系, 辽宁 大连 116024
  • 收稿日期:2021-02-23 出版日期:2022-01-01 发布日期:2022-01-19
  • 通讯作者: 刘洁
  • 作者简介:韩维(1970—), 男, 教授, 博士, 主要研究方向为飞行器动力学、舰载飞行器使用工程|崔凯凯(1991—), 男, 博士研究生, 主要研究方向为飞行器动力学、舰载机着舰控制|刘洁(1990—), 男, 助理研究员, 博士, 主要研究方向为舰载飞行器使用工程|王昕炜(1992—), 男, 讲师, 博士, 主要研究方向为最优控制|张勇(1978—), 男, 副教授, 硕士, 主要研究方向为舰载飞行器使用工程

Carrier landing control technology based on self-tuning MPC

Wei HAN1, Kaikai CUI1, Jie LIU2,*, Xinwei WANG3, Yong ZHANG1   

  1. 1. School of Basic Sciences for Aviation, Naval Aviation University, Yantai 264001, China
    2. War Research Institute, Academy of Military Sciences, Beijing 100850, China
    3. Department of Engineering Mechanics, Dalian University of Technology, Dalian 116024, China
  • Received:2021-02-23 Online:2022-01-01 Published:2022-01-19
  • Contact: Jie LIU

摘要:

针对模型参数不准确条件下的全自动着舰控制技术进行了研究, 设计了一种基于保辛伪谱算法(symplectic pseudospectral method, SP)和带遗忘因子递推最小二乘法(recursive least squares with forgetting factor, FFRLS)的舰载机着舰自校正模型预测控制方法(self-tuning model predictive control, ST-MPC)。针对舰载机的着舰控制模型, 在MPC方法的框架下, 首先设计了一种基于预测轨迹形状与位置偏差的引导轨迹,用以消除航母甲板运动和实时误差的影响。同时, 基于SP算法设计了滚动优化模块并引入雄鸡尾流扰动补偿以抑制舰尾流干扰; 进一步, 通过FFRLS算法结合巴特沃斯低通滤波器,对模型中影响控制性能的敏感参数进行实时估计用以提高着舰控制算法的鲁棒性。仿真结果表明, 在有多种干扰影响且控制模型参数不准确时, 所设计的ST-MPC算法, 能够将着舰高度误差控制在±0.15 m以内, 其控制精度远高于传统的线性二次型最优调节器和极点配置算法, 且算法的计算效率满足实时在线跟踪的要求。

关键词: 自校正, 保辛伪谱算法, 预测控制, 舰载机着舰, 参数不确定

Abstract:

In this paper, the carrier landing problem under the condition of model uncertainty is studied, and a self-tuning model predictive control method (ST-MPC) based on symplectic pseudospectral method (SP) and recursive least squares with forgetting factor (FFRLS) is established. Aiming at the carrier landing control model and in the framework of MPC method, a guidance trajectory based on the predicted shape and position deviation is designed to eliminate the influence of deck motion and real-time trajectory error. Meanwhile, the rock tail air wake compensation is introduced into the rolling optimization module based on SP algorithm to compensate the carrier air wake disturbance. Then, FFRLS with Butterworth low-pass filter is used to estimate the sensitive parameters in order to improve the robustness of the control algorithm. Simulation results show that the ST-MPC algorithm designed in this paper can control the height error within ±0.15 m with the disturbances are existed and the model parameters are not accurate, and its control accuracy is much higher than the traditional linear quadratic regulator and pole assignment algorithm. The calculation efficiency is also sufficient for real-time online tracking.

Key words: self-tuning, symplectic pseudospectral method, predictive control, carrier landing, parameter uncertainty

中图分类号: