引用本文:朱齐丹,王立鹏,张智,闻子侠.舰载机着舰侧回路时变风险权值矩阵线性变参数预测控制[J].控制理论与应用,2015,32(1):101~109.[点击复制]
ZHU Qi-dan,WANG Li-peng,ZHANG Zhi,WEN Zi-xia.Aircraft lateral linear parameter varying model predictive control with time varying weight[J].Control Theory and Technology,2015,32(1):101~109.[点击复制]
舰载机着舰侧回路时变风险权值矩阵线性变参数预测控制
Aircraft lateral linear parameter varying model predictive control with time varying weight
摘要点击 2291  全文点击 1218  投稿时间:2014-03-20  修订日期:2014-07-17
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2014.40224
  2015,32(1):101-109
中文关键词  自动着舰  横侧向控制  线性变参数  预测控制  线性矩阵不等式
英文关键词  automatic landing  lateral control  linear parameter-varying  model predictive control  linear matrix inequalities
基金项目  国家自然科学基金项目(61104037, 61304060), 国家国际科技合作专项基金(2013DFR10030), 中央高校基本科研业务费专项资金(HEUCF041307, HEUCFX41304), 中央高校基本科研业务费重大项目计划(HEUCFD1416)资助.
作者单位邮编
朱齐丹 哈尔滨工程大学 自动化学院 150001
王立鹏* 哈尔滨工程大学 自动化学院 
张智 哈尔滨工程大学 自动化学院 
闻子侠 哈尔滨工程大学 自动化学院 
中文摘要
      针对舰载机钩索阻拦阶段, 提出阻拦风险函数的概念和计算方法, 通过离线计算获得时变状态权值矩阵和时变控制输入权值矩阵. 采用时变状态权值矩阵可以实时调整各状态之间的变化关系, 而时变的控制输入权值矩阵能够调整副翼和方向舵的输入峰值, 避免出现输入饱和情况, 提高横侧向自动着舰的效率和安全性. 建立了时变的系统输出约束函数, 增加了舰载机进舰阶段控制器的可解性. 建立了基于状态偏差的舰载机横侧向着舰模型, 针对舰载机上很多状态不能直接测量的特点, 采用离线设计状态观测器的方法来估计舰载机的真实状态值, 利用线性矩阵不等式求解系统最优解, 通过舰载机自动着舰三维仿真模拟平台验证了算法的可行性.
英文摘要
      The concept and calculation method of arresting risk function are provided for the carrier aircraft hanging and arresting stage. Time-varying weight matrix of states and controlling inputs can be acquired by means of off-line calculation. The varied relation-ship between different states could be adjusted by employing time-varying states weight matrix in real time. At the same time, the controlling peak of aileron and rudder can be adjusted by varying controlling inputs weight matrix to avoid inputs saturation in real time. Therefore, the efficiency and safety of aircraft automatic landing could be improved. Time-varying system output constraint function is established to increase the feasibility of controller in the aircraft initial leading stage. Carrier aircraft lateral landing model is set up based on state deviations. For many states of the aircraft could not be measured directly, an off-line states observer is designed to estimate the real states of aircraft. Optimal solutions are solved by linear matrix inequalities. The simulation results verify the feasibility of the algorithm in the aircraft automatic landing three-dimensional simulation platform.