摘要
针对遭受外部扰动和输入饱和约束条件下的导弹自动驾驶仪系统,研究其鲁棒最优控制问题。首先,引入光滑函数处理输入饱和约束,并设计超螺旋干扰观测器实现对外部扰动的实时估计;然后,结合滑模控制理论,设计基于观测器的自适应滑模控制策略;接着,针对等效滑动模态系统,引入自适应动态规划算法构造评价网络求解哈密顿⁃雅可比⁃贝尔曼(HJB)方程,设计新的更新律在线学习最优控制策略使得系统稳定的同时满足预设的性能指标;最后,导弹自动驾驶仪系统的仿真结果表明了算法的有效性。
Aiming at a robust optimal control issue for missile autopilot systems with external disturbances and constrained input,firstly,a smooth function is employed to handle the constrained input and also a su⁃per⁃twisting disturbance observer is designed to estimate unknown external disturbance.Then,based on the sliding mode control theory,an observer⁃based adaptive sliding mode control policy is designed.Subsequent⁃ly,regarding equivalent sliding mode dynamic systems,an adaptive dynamic programming is introduced to establish a critic network to solve the Hamilton⁃Jacobi⁃Bellman(HJB)equation and a novel updating law is proposed to learn the optimal control strategy online,which makes the system stable and meets the prede⁃fined performance index.Finally,the missile autopilot system is used to indicate the effectiveness of the de⁃signed approach.
作者
赵锋
田辈辈
刘奇
Zhao Feng;Tian Beibei;Liu Qi(College of Mechanical and Electrical Engineering,Jiaozuo University,Jiaozuo 454000,China)
出处
《航天控制》
CSCD
北大核心
2023年第6期11-17,共7页
Aerospace Control
基金
河南省高等学校重点科研项目计划(23B210007)。
关键词
导弹
自适应动态规划
输入饱和
超螺旋观测器
自适应滑模控制
自动驾驶仪
Missile
Adaptive dynamic programming
Constrained input
Super⁃twisting disturbance ob⁃server
Adaptive sliding model control
Autopilot system
